List of posters

Poster session 10:30-12:00 - Result of students

Absorbčná spektroskopia radikálov produkovaných výbojom
Peter Čermák, Shamaila Manzoor, Filip Pastierovič
Abstract: Náš výskum sa zameriava na detekciu radikálov ako sú NO2 a NO3, monitorovanie ich časových zmien vo výbojoch a vývoj chemicko-kinetického modelu na lepšie pochopenie získaných údajov. NO2 a NO3 hrajú kľúčovú úlohu v atmosférickej chémii, najmä počas noci. Ich detekcia v plazmovom výboji, najmä NO3, je náročná vzhľadom na ich vysokú reaktivitu a nízku koncentráciu. Z aplikačného hľadiska sa javí zaujímavá hlavne tvorba plazmových zdrojov optimalizovaných na produkciu práve NO3. Preto sa v našom laboratóriu zameriavame na priamu detekciu týchto špecifických radikálov a možnosti optimalizácie plazmových zdrojov na produkciu NO3. Za týmto účelom sme zostrojili detektor na princípoch absorpčnej spektroskopie využívajúci optický rezonátor na zvýšenie citlivosti zariadenia (CEAS). Špecifikom tohto detektora je využitie širokospektrálneho a nekoherentného zdroja svetla (IBB-CEAS), čo taktiež zabezpečuje dostatočnú robustnosť detektora, potrebnú pri práci s plazmou.
Alignment-free detection of copy number variants using k-mer counts
Alessia Petescia, Askar Gafurov, Tomas Vinar, Brona Brejova, Luca Denti
Abstract: Variant calling strategies aim to identify genetic differences between a sequenced genome and a reference genome. Traditional methods relying on read mapping can struggle with highly divergent genomes or when a reliable reference assembly is not available. At the same time, those methods are computationally intensive. We present an alignment-free, k-mer-based method for detecting copy number variants (CNVs). By comparing k-mer counts between two samples, our method delivers results comparable to state-of-the-art tools while being faster and more reliable, particularly in regions with segmental duplications.
All McKay-Miller-Širáň graphs are bi-Cayley
Dávid Wilsch, Martin Mačaj
Abstract: Inspired by the degree-diameter problem, McKay, Miller and Širáň introduced a family of large and highly symmetric graphs $H_q$, where $q$ is a prime power greater than two. These graphs are of diameter $2$, order $2q^2$ and degree $(3q-r)/2$, where $q$ is congruent to $r$ mod 4. Moreover, McKay-Miller-Širáň (MMS) graphs $H_q$ with $q=3$, $q=4$ and $q=4k+1$ are vertex transitive but not Cayley and for other values of $q$ the automorphism group splits the vertices into two orbits.
Analysis of elemental composition of atmospheric aerosols in Bratislava using Particle Induced X-ray Emission (PIXE) method with an external beam
Karol Sučák , Miroslav Ješkovský, Jakub Kaizer, Ivan Kontuľ, Pavel P. Povinec, Jakub Zeman
Abstract: In addition to gases, the atmosphere is filled with small solid or liquid particles called aerosols. The study of these particles is important for climate change and environmental research, as aerosols are an important pollutant and may have adverse effects on human health. The different properties of aerosols are being studied in laboratories around the world. In this work, the PIXE method was used to study the elemental composition of aerosol. Aerosol samples were collected at the meteorological station on the campus of the Faculty of Mathematics, Physics, and Informatics in Bratislava. Quartz microfiber filters are used to collect particulate matter, while approximately 1500 m^3 of air is pumped through each filter. The samples were collected in two weeks interval over a year between April 2023 and April 2024. The total mass and composition of the collected aerosols were determined by gravimetric analysis and the PIXE method, respectively. The Centre for Nuclear Technology and Accelerator Technology (CENTA) operates a 3 MV Pelletron accelerator (NEC, United States) with the necessary infrastructure for the production and extraction of selected ion beams. In 2024, a new external beam line was set up with a SiPIN detector from AMPTEC Inc. (USA) for in-air PIXE analysis. PIXE analysis of the samples identified nine elements in atmospheric aerosols, where Si, S, K, Ca, and Fe were the most abundant. Since quartz filters are used for sampling and silicon detectors are used for analysis, the observed Si concentrations are only indicative.
Anomalous R-T Characteristics of Superconducting NbN Thin Films
Magdaléna Poláčková, Elena Zhitlukhina, Mikhail Belogolovskii, Tomáš Plecenik, Maroš Gregor
Abstract: The resistance-versus-temperature curve of superconductors is typically monotonically decreasing at critical temperature Tc. However, some measurements performed on the superconducting NbN layer by the van der Pauw method showed a resistance peak just before Tc which disappeared with the rotation of the contacts. Thus, the 50 nm thick NbN layer was analyzed in detail. The sample was prepared by pulsed laser deposition on a c-cut sapphire substrate with a 50 nm thick AlN buffer layer. Its critical temperature was 13.4 K. The contact rotation led to the critical temperature increase by the peak width to 13.8 K. The anomalous behavior was explained as a result of the superconducting inhomogeneity of the layer since the superconducting granularity could lead to slightly different values of Tc and ΔTc of the different film sections. Experimentally measured data were fitted by a theoretical model, which enabled the determination of values of Tc and ΔTc of individual film sections. Furthermore, a perpendicular magnetic field of 6 Tesla led to the minimization of the difference between two contact configurations and to the disappearance of the peak. This result could be a consequence of the global averaging of the order parameter.
Application of Cold Plasma to pesticide degradation
Angel Pardo, Matej Klas, Štefan Matejčík
Abstract: According to World Health Organization estimates, the global population is expected to increase by around 20 percent by 2050, requiring an increase in crop productivity to meet food demand, therefore the use of pesticides cannot be reduced despite their health risks, and they will continue to play a critical role in preventing crop losses by pests and diseases. After applying pesticides to crops, some may be found in food at a residue level due to their low solubility in water and stability under environmental conditions, making them resistant to degradation. Currently, one of the challenges is to develop technologies that effectively eliminate pesticide residues without causing damage to the nutritional content of the food. Cold plasma technology is a promising tool for pesticide removal because of its ability to degrade organic molecules through reactive oxygen and nitrogen species, radicals, and UV radiation that compose the plasma. Evaluating the degradation of chlorpyrifos using Corona Discharge is the aim of this study. The use of Chlorpyrifos in the EU has been banned since 2020 and was selected for this study based on its repeated appearance in the last results obtained by The Multiannual National Control Program to monitor the pesticide residue carried out by the European Food Safety Authority, also was consider its risk to human health due the effect of developmental neurotoxicity and genotoxic. The quantification of degradation percentage and identification of degradation products was made by Ion Mobility Spectrometry and Mass Spectrometer. The 3,5,6-trichloro-2-pyridinol was the main product of the degradation by Corona Discharge.
Automatic Classification and Detection of white matter lesions in multiple sclerosis patients' MRI images using Deep Learning and Explainable AI Methods
Fatana Jafari
Abstract: White Mater Lesions in MS is an autoimmune disease that affects the body's central nervous system. This illness results in severe symptoms for the body, like muscle imbalance and blurred vision. Even though AI-assisted multiple sclerosis or white matter lesion identification systems have proven to be quite successful, their adoption in clinical settings has encountered many challenges, mostly due to the limited explainability of their fundamental principles. Physicians are cautious regarding AI-assisted diagnoses due to the lack of ability to be transparent about decision-making procedures. In this regard, the computational black box issue with AI systems has been resolved by explainable artificial intelligence (XAI), which was developed to provide explanations for predictions made by models. The primary objective of this research has been to explore XAI techniques such as Shapley additive explanations (SHAP) and local interpretable model-agnostic explanations (LIME), which facilitate the explanation of model predictions. In addition, we use various CNN models to train the data of brain MRI images with and without white matter lesions, which are collected from the radiology department at Cyril and Methodius Hospital in Bratislava.
Biaxiálna textúra v žíhaných TiBx vrstvách deponovaných na Al2O3 (0001)
R. Vrablec, B. Grančič, T. Roch, V. Šroba, V. Izai, M. Truchlý, T. Fiantok, M. Mikula
Poster
Abstract: Boridy prechodných kovov s pomerom bóru ku kovu v rozmedzí 1 až 2 tvoria štruktúrne zaujímavú skupinu materiálov s množstvom stabilných kryštalických fáz. Zároveň sa v týchto materiáloch vyskytujú rôzne plošné defekty, ktoré umožňujú začlenenie prebytočného kovu/bóru, dvojčatenie alebo koherentnú väzbu dvoch rôznych fáz [1]. V našom príspevku skúmame štruktúru tenkých vrstiev TiB (pomer bóru ku kovu 1:1) nanesených na substrátoch Al2O3 (0001) naprašovacou technikou High Target Utilization Sputtering (HiTUS) a ex-situ vákuovo žíhaných do 1200°C. Opisujeme vzájomnú orientáciu všetkých kryštalických fáz prítomných v povlaku vzhľadom na substrát pomocou röntgenovej difrakčnej analýzy, konkrétne meraním pólových obrazcov a mapovania recipročného priestoru. Referencie [1] K. Viskupová, B. Grančič, T. Roch, Š. Nagy, L. Satrapinskyy, V. Šroba, M. Truchlý, J. Šilha, P. Kúš, M. Mikula, Thermally induced planar defect formation in sputtered V1-xMoxB2-Δ films, Scripta Materialia, 229 (2023) 115365, https://doi.org/10.1016/j.scriptamat.2023.115365
Cognitively-inspired learning of causal relations in a simulated robotic environment
Miroslav Cibula, Igor Farkaš
Poster
Abstract: Learning causal relations allows humans to predict the effect of their actions on the known environment and use this knowledge to plan the execution of more complex actions. Such knowledge also captures the behaviour of the environment and can be used for its analysis and the reasoning behind the behaviour. In this thesis, we explore learning causal relations observed in a simulated robotic environment using the forward and inverse models. Inspired by mechanisms of human cognition, we propose multiple approaches to constructing such models. Further, we investigate how these models trained on synthetic data generated in a simulation can be analyzed to extract learned low-level causal relationships, which could be then used as a basis for dimensional reduction of the environment's state representations as well as for the explainability of the environment's behaviour at higher levels. Finally, we propose an approach for planning actions for simple robotic manipulation task-solving using some of the developed methods and concepts presented in this work. In order to evaluate the proposed methods, we conduct several experiments concerning a simulated robotic arm learning causal relations in tasks of increasing difficulty.
Cold Atmospheric Plasma for Mould Control
Zuzana Okruhlicová, Karol Hensel
Abstract: The susceptibility of fresh produce to microbial spoilage presents a major challenge for agriculture and the food industry, leading to substantial economic losses and public health risks globally. Among citrus fruits, Penicillium digitatum is a key postharvest phytopathogen, responsible for visible green mould. Conventional chemical fungicides, though widely used, raise environmental and health concerns and contribute to increasing antifungal resistance. This has created an urgent need for sustainable alternatives to enhance food safety and extend the shelf life of citrus fruits. Cold atmospheric plasma (CAP) technology is emerging as a promising and eco-friendly solution for fungal inactivation. In this study, we examined two CAP methods—DC transient spark (TS) discharge and pulsed streamer corona (SC) discharge—to inactivate P. digitatum spores. Two treatment approaches were assessed: direct plasma treatment of spore suspensions and indirect treatment via plasma-activated water (PAW), which was subsequently applied to spores or infected lemons. The efficacy of PAW was also compared with that of low-pH water. Additionally, we evaluated a delayed sporicidal effect by inoculating treated spore suspensions onto agar plates 30 minutes post-treatment. The physicochemical properties of PAW, including pH, conductivity, and reactive oxygen and nitrogen species (e.g., H₂O₂ and NO₂⁻), were also analysed. Funded by the EU NextGenerationEU through the Recovery and Resilience Plan for Slovakia under the project No. 09I03-03-V03-00033.
Combined Antimicrobial Properties of Ultraviolet Radiation and Plasma-Activated Water
Ramin Mehrabifard, Bernard Gitura, Zdenko Machala
Abstract: Non-equilibrium plasma discharges at atmospheric pressure are used in microelectronics, polymer deposition, light sources, medicinal applications, and environmental remediation [1–3]. Biomedical uses of cold atmospheric plasma produced in ambient air are emerging due to their unique features. Both long-lived and short-lived reactive oxygen and nitrogen species (RONS) are produced by non-equilibrium plasma in the air atmosphere. These long-lived RONS are transported into plasma-activated water (PAW) solutions, which are produced when atmospheric plasmas and water come into contact. PAW solutions are efficient for both killing microorganisms and inactivating cancer cells. On the other hand, UV radiation (UVA 315-400 nm, UVB 280-315 nm, UVC 120-280 nm) is commonly employed for water and air purification, medical sterilization, food and beverage sector, and analytical methods [4–6]. The goal of this research is to better understand how cold air plasma interacts with bacteria by examining the impact of UVA radiation as a supporting factor on the production of specific RONS in gas and water, as well as the subsequent effects on bacteria viability. The investigation specifically examined RONS produced by transient spark (TS) discharge plasma and UV radiation in both the gaseous and liquid phases upon the interaction between plasma and water. The concentrations of the gaseous plasma species, namely H2O2, O3, and (H)NOx (HNO2, NO2, and NO), as well as their dissolution in water as H2O2(aq), O3(aq), nitrites NO2¯(aq), nitrates NO3¯(aq) are measured. Figure 1 shows the schematic diagram of our UV/ plasma setup. Different modes of treatment are examined including plasma alone, UV alone, plasma and UV, plasma followed by UV, and reverse.
Fig. 1 schematic of UV/plasma setup Acknowledgment: This work was supported by Slovak Research and Development Agency APVV-22-0247 and Comenius University Grant UK/3032/2024. Funded by the EU NextGenerationEU through the Recovery and Resilience Plan for Slovakia under the project No. 09I03-03-V03-00033.
References [1] Ramoy et al. J. Phys. D. Appl. Phys., vol. 57, no. 1, p. 01LT01, 2024 [2] Friedrich et al. Nonthermal Plasmas Mater. Process., 2022 [3] T. Il Lee at al. Appl. Phys. Lett., vol. 87, no. 26, pp. 1–3, 2005 [4] A. Berardinelli et al., Chemosphere, vol. 262, 2021 [5] E. M. Plein, Am. J. Heal. Pharm., vol. 34, no. 12, pp. 1372–1372, 1977. [6] R. D. Tchonkouang. Foods, vol. 12, no. 17, 2023
CV-Probes: Studying the interplay of lexical and world knowledge in visually grounded verb understanding
Ivana Beňová, Michal Gregor, Albert Gatt
Abstract: This study investigates the ability of various vision-language (VL) models to ground context-dependent and non-context-dependent verb phrases. To do that, we introduce the CV-Probes dataset, designed explicitly for studying context understanding, containing image-caption pairs with context-dependent verbs (e.g., ""beg"") and non-context-dependent verbs (e.g., ""sit""). We employ the MM-SHAP evaluation to assess the contribution of verb tokens towards model predictions. Our results indicate that VL models struggle to ground context-dependent verb phrases effectively. These findings highlight the challenges in training VL models to integrate context accurately, suggesting a need for improved methodologies in VL model training and evaluation.
Diffusing in Someone Else’s Shoes: Robotic Perspective-Taking with Diffusion
Matthias Kerzel, Stefan Wermter
Abstract: Humanoid robots can benefit from their similarity to the human shape by learning from humans. When hu- mans teach other humans how to perform actions, they often demonstrate the actions, and the learning human imitates the demonstration to get an idea of how to perform the action. Being able to mentally transfer from a demonstration seen from a third-person perspective to how it should look from a first-person perspective is fundamental for this ability in humans. As this is a challenging task, it is often simplified for robots by creating demonstrations from the first-person perspective. Creating these demonstrations allows for an easier imitation but requires more effort. Therefore, we introduce a novel diffusion model that enables the robot to learn from the third-person demonstrations directly by learning to generate the first-person perspective from the third-person perspective. The model translates the size and rotations of objects and the environment between the two perspectives. This allows us to utilise the benefits of easy-to-produce third-person demonstrations and easy-to-imitate first-person demonstrations. Our approach significantly outperforms other image-to-image models in this task.
Docker Survey for FLOps Efficiency
Ladislav Hluchý, Peter Krammer
Abstract: This paper presents a comprehensive analysis of Federated Learning (FL) deployment using Docker containers, termed Docker-enabled Federated Learning (DFL), in High- Performance Computing (HPC) environments. We emphasise the seamless deployment of FL through Docker, achieving scalability, reduced hardware resource usage, privacy protection, and IoT compatibility. An extensive literature review was conducted on the efficiency, memory usage, and CPU consumption under varying concurrent client loads. Findings indicate that DFL efficiently handles multiple simultaneous clients while maintaining satisfactory average accuracy. DFL protects privacy by ensuring that only clients can access their generated data streams, enabling the development of a robust global model without compromising data privacy.
Electron Impact Excitations of Astrophysically Relevant Molecules
Enmily García, Barbora Stachová, Juraj Országh, Štefan Matejčik
Abstract: Electron–molecule interactions play a critical role across various fields. Understanding the kinetics of these interactions and their resulting products has enabled the development of thin-film deposition techniques, such as Focused Electron Beam Induced Deposition (FEBID), as well as discharge applications in plasma medicine, material processing, or water purification. Emission spectroscopy is an analytical method allowing the study of electron-molecule collisions producing excited species, which is especially important for astrophysics. Emission spectroscopy is particularly valuable in space exploration, as it enables remote analysis of planetary atmospheres, cometary comas, and nebulas irradiated mainly by nearby stars. The secondary electrons generated in photoionization reactions further interact with gas molecules and emission generated through these reactions provide insights into the molecular composition and processes active in these interstellar bodies. In general, there is a serious lack of kinetics data for the electron-molecule reactions. At the Department of Experimental Physics, the Laboratory of Electron Induced Fluorescence (LEIF) houses an experimental setup designed to investigate these radiative processes. This research provides essential reference data for analyzing astronomical spectra, supports theoretical modeling of complex systems (e.g., atmospheric processes and electric discharges), and underpins various research and industrial applications.
Electron Impact Excitation of Carbon Monoxide
B. Stachová, J. Blaško, J. Országh, S. Bromley, D. Bodewits, Š. Matejčík
Abstract: Carbon monoxide is one of the dominant carbon bearing molecules in extra-terrestrial bodies such as comets or centaurs. Fluorescent radiation of CO and CO+ is especially prominent in the emission spectra of comet tails. The diagnostic of these cometary volatiles is a necessity for solar system formation models. Electron-induced fluorescence of CO was studied in a crossed-beam experiment. The emission spectrum composed of neutral CO and CO+ emission bands, and atomic lines of C and O, was measured at several electron energies ranging from 5 to 100 eV within the wavelengths of 280 - 1030 nm. Excitation-emission functions for all identified transitions of the spectrum were measured as well and their threshold energies were estimated.
Elemental analysis of selected wines by Particle Induced X-ray Emission (PIXE) for determination of geographical origin
Boris Bobáľ, Miroslav Ješkovský, Jakub Kaizer, Ivan Kontuľ, Marek Bujdoš, Pavel P. Povinec
Abstract: Varietal wines are recognized worldwide for their taste and quality, produced by numerous wineries adhering to protected appellations and traditional production processes. This study applies multi-elemental techniques and statistical tools to authenticate the geographical origin of wines and verify compliance with international and national regulatory concentration limits. The analysis encompasses 21 wine samples from five countries on four continents, including Slovak wines from notable vineyard regions such as Chardonnay (white) and Frankovka modrá (red). Furthermore, samples from Australia (Sauvignon Blanc), Argentina (Cabernet Sauvignon), New Zealand (Sauvignon Blanc), and South Africa (Chenin Blanc) were selected, with three samples each of consistent terroir, vintage, and variety. Each 250 ml sample was analysed, incorporating Yttrium as an internal standard to ensure concentration accuracy. The samples were initially evaporated at 80°C, followed by ashing at 650°C for 15 hours. The resulting ash was pulverised and pressed into pellets for Particle-induced X-ray emission (PIXE) analysis. Conducted in the CENTA laboratory, the measurements utilised a 3 MV NEC linear tandem accelerator with 3 MeV proton beams to detect 17 elements. To validate the results, a subset of liquid samples was subjected to inductively coupled plasma mass spectrometry (ICP-MS) analysis. Elements identified included Na, Mg, Al, Si, P, S, Cl, K, Ca, Mn, Fe, Cu, Zn, Br, Rb, Sr, and Ba, which were used in statistical analyses to determine the geographical origin of each sample. A sample preparation methodology was developed to quantify trace elements, comparing concentrations obtained from PIXE and ICP-MS, with origins validated through linear discriminant analysis (LDA).
ELMiRA: Embodying Language Models in Robot Action
Ozan Özdemir, Cornelius Weber, Stefan Wermter
Abstract: Large language models (LLMs) have achieved significant recent success in deep learning. The remaining challenges in robotics and human-robot interaction (HRI) still need to be tackled but off-the-shelf pre-trained LLMs with advanced language and reasoning capabilities can provide solutions to problems in the field. In this work, we realise an open-ended HRI scenario involving a humanoid robot communicating with a human while performing robotic object manipulation tasks at a table. To this end, we combine pre-trained general models of speech recognition, vision-language, text-to-speech and open-world object detection with robot-specific models of visuospatial coordinate transfer and inverse kinematics, as well as a task-specific motion model. Our experiments reveal robust performance by the language model in accurately selecting the task mode and by the whole model in correctly executing actions during open-ended dialogue. Our innovative architecture enables a seamless integration of open-ended dialogue, scene description, open-world object detection and action execution. It is promising as a modular solution for diverse robotic platforms and HRI scenarios.
Enhancement of the gasistor performance by separating the memristive and gas sensing cell
Michal Patrnčiak, Ľubomír Staňo, Ihor Shpetnyy, Marek Vidiš, Martin Moško, Tomáš Roch, Maroš Gregor, Tomáš Plecenik
Abstract: The gasistor device combines chemiresistive gas sensing and resistive switching in a single capacitor-like metal/metal oxide/metal cell. In previous works, the same cell was used for both gas-sensing and resistive switching. The mutual entanglement of their fabrication parameters impedes the ability to adjust the two functionalities independently. After their separation, we can inspect the memristive cell and adjust the fabrication parameters, so the cell is more compatible with the gas sensor, potentially improving the reliability and sensitivity of the gasistor device.
Exploring Pre-Service Teachers' Ideas and Beliefs Towards Reasoning and Proof in Mathematics and Informatics Education
Katarína Jánošková
Abstract: This study explores the beliefs of Slovak pre-service teachers regarding the use of reasoning and proof (R&P) in mathematics and informatics education, as emphasized by the national curriculum. Group interviews were conducted with first-year master’s students in teacher training programs for these subjects. Results indicate similarities in perceptions of R&P across the two groups, with respondents noting challenges in incorporating R&P and identifying specific topics where they plan to apply it in their future teaching. This research is part of a larger study within the MaTeK project, aimed at tracking changes in students' beliefs over the course of their studies.
Exploring the effect of key parameters in atmospheric-pressure plasma polymerization for layer deposition
Sandra Ďurčányová, Michal Stano, Pedram Ghourchi Beigi, Dušan Kováčik
Poster
Abstract: Preparation of thin, functional layers is an important process in various industries. Plasma polymerization (PP) enables deposition of such layers by using a plasma source. The properties of PP-layers are affected by a multitude of working conditions. Despite extensive research, the specific effects of each parameter on layer properties remain unclear. This contribution deals with the study of layers deposited by PP of hexamethyldisiloxane (HMDSO) using a Diffuse Coplanar Surface Barrier Discharge (DCSBD) at atmospheric pressure. Our objective is to investigate how changes in working conditions impact the functionality, thickness and chemical composition of the deposited layers. Surface diagnostics such as WCA, SEM and ATR-FTIR were employed to characterize the prepared layers.
Heterostructural decomposition in hard V1-xWxB2-delta films prepared by magnetron sputtering
Katarína Viskupová, Tomáš Fiantok, Branislav Grančič, Peter Švec ml., Tomáš Roch, Martin Truchlý, Viktor Šroba, Leonid Satrapinskyy, Marián Mikula
Abstract:
Intensity calibration of the broadband echelle type spectrometer for the precise LIBS spectra
Vaishnavi Sudhakaran, Sanath Shetty, Pavel Veis
Abstract: In Laser-induced breakdown spectroscopy (LIBS), plasma is produced when a high-intensity laser pulse reaches the sample surface which is then ablated and dissociated to the single particles, which are consequently excited and ionized to form the so-called laser induced plasma (LIP). Later, the local thermodynamic equilibrium is formed and excited states are radiatively decaying by emission forming atomic optical emission LIBS spectrum. As the LIBS spectrum is unique the scientific preference is to captured whole spectrum from UV to NIR range in one acquisition with broadband spectrometer equipped with an iCCD camera. The raw intensity recorded by spectrometers can be influenced mainly by detectors (thermal and read out noise) and instrument spectral responses that depend on wavelength. Using the special echelle based broadband spectrometer the spectrum is recorded through many diffraction orders doing the spectral response curve much more complicated with oscillating structure. The intensity calibration becomes of utmost importance to obtain accurate quantitative information from spectral data. Calibration corrects these inconsistencies and allows uniform spectral responses throughout the spectrum. In this work, we have improved the standard spectral response sensitivity evaluation [1] by performing the necessary mathematical operations to smooth existing responses, thereby achieving more accurate LIBS signals. References: [1] Rakovský, J., et al Measurement of echelle spectrometer spectral response in UV, In: WDS 2011: Proceedings of Contributed Papers: Part 2 Physics of Plasmas and Ionized Media. - Prague : MATFYZPRES, 2011, pp. 257-262, ISBN 978-80-7378-185-9. Acknowledgement: The authors acknowledge the Scientific Grant Agency of the Slovak Republic (contract number VEGA 1/0803/21) and the Slovak Research and Development Agency (APVV-20-0111, APVV-22-0548, and APVV-23-0281) for financial support.
Invisible Spectrum Quotient: Moving Toward the Unfolding of Socially High-Functioning Asperger Syndrome
Alexandra Dyalee
Abstract: Asperger syndrome is a neurotype characterized by difficulties in social interaction, accompanied by a strong intrinsic motivation present within an individual’s unique patterns of interest. Despite its expectedly noticeable behavioral manifestations, in a number of cases, Asperger’s can be difficult to get properly identified, which results in frequent instances of late-adulthood diagnosis accompanied by long-term mistreatment. The reason for the latter phenomenon is the fact that despite having a formal diagnosis of Asperger syndrome, some individuals are paradoxically often able to perform successfully in demanding social situations. Most commonly, this is a case of females. With an aim to minimize this identification error in society, I am developing a novel method for diagnostics of Asperger syndrome in socially high-functioning individuals based on the correlation of social functioning and cognitive load, which has been shown to be a side effect of different brain regions involved in successful realization of ToM tasks in these individuals.
Josephson Traveling Wave Parametric Amplifiers with Plasma oscillation phase-matching
Emil Rizvanov, Samuel Kern, Pavol Neilinger and Miroslav Grajcar
Abstract: High gain and large bandwidth of traveling-wave parametric amplifier exploiting the nonlinearity of Josephson Junctions can be achieved by fulfilling the so-called phase-matching condition. This condition is usually addressed by placing resonant structures along the waveguide or by periodic modulations of its parameters, creating gaps in the waveguide’s dispersion. Here, we propose to employ the Josephson junctions, which constitute the centerline of the amplifier, as resonant elements for phase matching. By numerical simulations in JoSIM (and WRspice) software, we show that Josephson plasma oscillations can be utilized to create wave vector mismatch sufficient for phase matching as well as to prevent the conversion of the pump energy to higher harmonics. The proposed TWPA design has a gain of 15 dB and a 3 GHz bandwidth, which is comparable to the state-of-the-art TWPAs.
Leveraging Meta-Modelling Language for Ontology Structuring and Validation.
Zekeri Adams, Martin Homola, Ján Kľuka, Vojtěch Svátek
Abstract: Knowledge representation (KR) is essential for AI, providing the formal structures needed for systems to reason and understand complex domains. It underpins applications like expert systems, intelligent agents, and robotics. However, traditional KR methods struggle with the evolving complexity in fields like biomedicine, finance, and cybersecurity, where rapid growth and interdependencies pose challenges to accurate representation. Meta-modeling offers a flexible solution by introducing reusable frameworks that improve scalability and adaptability. Our work focused on the formalization of the meta-modeling languages, PURO, which incorporate higher-order features to enhance ontology clarity and semantic richness. We explore how this formalization can improve the organization and precision of knowledge representation, with practical application in Wikidata.
LIBS in experiments in liquids
Adam Nagy, Jairo C. Peralta, Sanath Shetty, Pavel Veis
Abstract: Monitoring the chemical composition of liquids is essential for public health, environmental safety, and industrial regulation, particularly when it comes to detecting toxic heavy metals. Traditional methods for liquid analysis using Laser-Induced Breakdown Spectroscopy (LIBS) face difficulties due to the splash and evaporation of samples, leading to energy loss and inconsistent results. By combining LIBS with acoustic levitation (LIBS-Lev), we address these challenges, as levitation enables the suspension of droplets in a stable, contact-free environment, thereby preserving sample integrity and enhancing precision. LIBS-Lev provides rapid, real-time elemental analysis, making it highly suitable for detecting pollutants like lead, cadmium, and mercury in various liquid samples, from natural waters to industrial effluents. This novel approach allows for a broader range of applications in environmental and health monitoring, with the potential to expand its accuracy and sensitivity further in future research. LIBS-Lev stands out as a promising method for effective, non-invasive heavy metal monitoring, crucial for safeguarding water quality and supporting sustainability efforts.
Manifolds homotopy equivalent to trivial disk bundles over CP(n)
Samuel Kalužný
Abstract: This project is concerned with the problem of classifying different smooth structures admitted by a smooth manifold up to homotopy. This classification is made possible by the tools of surgery theory and is summarized in the notion of the smooth structure set S^{DIFF}(X) of a smooth manifold X. We focus on cases when X is the n-dimensional complex projective space and when X is the product of the complex projective space with a k-dimensional disk. In the first case we extend G. Brumfiel's computations of the structure sets of CP(2n) to dimensions 2n<20. In the second case we compute structure sets of products CP(n)xD^k for dimensions satisfying 2n+k=0 mod 4. We also determine which topological manifolds homotopy equivalent to CP(2n)xD^4, n<10, admit a smooth structure by looking at the map from the smooth structure set S^{DIFF}(CP(n)xD^4) to the topological structure set S^{TOP}(CP(n)xD^4).
Measures of edge uncoverability of cubic graphs
Robert Lukoťka, M. F. Oguagbaka
Poster
Abstract: Cubic graphs without bridges can be categorized in terms of their perfect matching index—the minimum number of perfect matchings needed to cover the edges of a bridgeless cubic graph. A conjecture of Berge suggests that at most five perfect matchings are needed to cover any such graph. Cubic graphs that need a minimum of five perfect matchings to cover their edges are very rare, and only a few infinite families of nontrivial examples of such graphs are known. Such graphs are of particular interest, as famous, longstanding open problems, including the cycle double cover conjecture, can be reduced to them. In this study, we define the four perfect matching cover defect of a cubic graph as the minimum number of edges of the graph not covered by four perfect matchings. This invariant serves as a way to capture how far a cubic graph is from being coverable by four perfect matchings. We also introduce several other invariants that capture this notion, and we refer to these invariants as measures of uncoverability of cubic graphs. We present an infinite family of cubic graphs that are cyclically 4-edge-connected, have girth at least 5, and are far from being coverable by four perfect matchings with respect to the invariants. On the other hand, we also present graphs where values of the invariants differ. We investigate how the introduced measures relate to well-known conjectures, particularly the 5-cycle double cover conjecture.
Meteor observations as a tool to constrain cosmogonic models of the Solar System
Vitalii Kuksenko, Juraj Tóth
Abstract: Cosmogonic models of the Solar System describe its origin and evolution involving various scenarios that are often based on different physical mechanisms. Recent observations of small bodies showed evidence of the presence of refractory objects in the Oort cloud, and different models produce different predictions of their numbers. The measurement of this population can be used as an observational constraint for cosmogonic models. The aim of our work was to study how the data obtained from meteor observations can be used as a tool for distinguishing among the existing cosmogonic models. In this poster, we will first review the cosmogonic models that we want to compare. We will investigate the main physical properties of meteoroids focusing on their material and orbital classifications. Then we will present the analysis of two meteor databases collected by the cameras of the All-Sky Meteor Orbit System (AMOS) located in the Canary Islands and in Chile. We will describe the methodology of the search of unusually strong rocky meteoroids on cometary orbits with the origin in the Oort cloud and present the results on the found rocky and cometary meteoroids with mass limits of 10 g and 1 g. These data in combination with the data from other authors were used to calculate the fluxes of meteors of different compositions. We used them to constrain the ratio of icy and rocky components of the Oort cloud. The comparison of the obtained results with other authors and with the predictions of different cosmogonic models will be presented.
Modelling the optical absorption of meandering niobium nitride nanowires in the infrared spectra
M. Baránek, P. Neilinger, S. Kern, M. Poláčková, M. Grajcar
Poster
Abstract: The State-of-the-art detectors enabling quantum key distribution are the superconducting nanowire single-photon detectors (SNSPD). The photon absorption efficiency, given by geometry of the nanowire and its optical conductivity, contributes to the overall detection efficiency. The optical conductivity of disordered metallic films, e.g. niobium nitride, is strongly affected by the presence of quantum corrections[3], resulting in its nontrivial wavelength and thickness dependence. We model the absorption spectra of NbN nanowires in a λ/4 resonator for different film thicknesses and show that the optical conductivity of thin films affects the absorption spectra. We show that the real part of the conductivity determines the amplitude of the absorption and the imaginary part contributes mostly to the wavelength shift of the maxima. This should be accounted for in detector design and optimization of its optical absorption
Non-linear refinement schemes for curves
Klaudia Hamajová, Pavel Chalmovianský
Poster
Abstract: Refinement schemes are used to obtain a better approximation of a particular curve using local refinement rules. These are techniques that allow the generation of smooth curves from discrete data through iterative refinement. We focus on non-linear planar refinement schemes for polygon lines where normal vectors are assigned at each vertex. These polygonal lines are subsequently refined using locally calculated conics. We present examples of the application of these schemes and analyze their properties.
Nonthermal Plasma Regeneration of Deactivated Catalysts
Jana Kšanová, Richard Cimerman, Christian Oberste-Beulmann, Leonid Satrapinskyy, Peter Švec, Karol Hensel
Abstract: The combination of nonthermal plasma (NTP) with catalysis, known as plasma catalysis, is extensively employed in various environmental applications, such as air pollutant removal. However, the catalysts deactivate over time, which limits their lifetime and requires regular regeneration or replacement. Since replacing catalysts can be costly, there is growing interest in finding efficient regeneration methods to extend catalysts lifetime and improve thier economic profitability.
Our research is therefore focused on the potential of NTP generated by packed-bed dielectric barrier discharge (PB-DBD) reactor at atmospheric pressure to regenerate deactivated catalysts. The experimental procedure involved two stages: first, a model pollutant (toluene) was removed from the air using the PB-DBD reactor. The process resulted in catalyst deactivation due to gradual accumulation of solid compounds on its surface. In the second stage, the deactivated catalyst was regenerated in the same PB-DBD reactor using pure oxygen. Different catalysts (TiO2, Pt/γAl2O3 and Pd/γAl2O3) and one non-catalytic material were tested (Al2O3). The gaseous products of the toluene removal as well as the products of the regeneration were monitored using infrared absorption spectroscopy (FTIR). The efficiency of plasma regeneration was compared with other methods such as ozone and thermal regeneration. Additionally, a sequential plasma regeneration method, in which the catalysts were mixed after a certain regeneration period, showed the highest efficiency.
To evaluate regeneration efficiency, the time courses of concentrations of gaseous products of regeneration (CO2, CO, HCOOH), thermogravimetric analysis (TGA) and scanning electron microscopy (SEM) were used. Furthermore, solid compounds on the catalyst surfaces were analyzed using gas chromatography coupled with mass spectrometry (GC-MS).
This work was supported by Excellent Grant of Comenius University Bratislava UK/3061/2024 and funded by the EU NextGenerationEU through the Recovery and Resilience Plan for Slovakia under the project No. 09I03-03-V04-00092 and the project No. 09I03-03-V03-00033.
Optical and transport properties of NbN thin films revisited
P. Neilinger, M. Poláčková, M. Baránek, T. Plecenik, T. Roch, M. Grajcar
Abstract: Highly disordered NbN thin films exhibit promising superconducting and optical properties. Despite extensive study, discrepancies in its basic electronic properties persist. Analysis of the optical conductivity of disordered ultra-thin NbN films, obtained from spectroscopic ellipsometry by standard Drude-Lorentz model, provides inconsistent parameters. We argue that this discrepancy arise from neglecting the presence of quantum corrections to conductivity in the IR range. To resolve this matter, we propose a modification to the Drude-Lorentz model, incorporating quantum corrections. The parameters obtained from the modified model are consistent not only with transport and superconducting measurements but also with ab initio calculations. The revisited values describing conduction electrons, which differ significantly from commonly adopted ones, are the electron relaxation rate Γ ≈ 1.8 eV/ℏ, the Fermi velocity vF ≈ 0.7 × 106 ms−1 and the electron density of states N(EF ) = 2 states of both spins/eV/NbN.
Optical diagnostics of transient spark with electrospray
P.Tóth, P.Pareek, S.Smirnov, M.Janda
Abstract: Transient spark (TS) produces high concentrations of NO, NO2, and HNO2. Solvation of these gas species produces NO2-, NO3-, H2O2, and ONOO-/ONOOH in the plasma activated water (PAW). The solvation of gas species can be enhanced by converting bulk water into electrospray (ES) microdroplets directly inside TS discharge zone. The simultaneous generation of TS and ES increases the concentration of iron ions from the sputtering of the electrodes. The iron ions can interact with H2O2 in the PAW by to Fenton reactions, potentially enhancing the biocidal properties of PAW. Funded by the EU NextGenerationEU through the Recovery and Resilience Plan for Slovakia under the project No. 09I03-03-V03-00033.
Optimising Federated Learning Multi-Task and Multi-Objective Approaches
Ladislav Hluchý, Lenka Skovajsová
Abstract: In the last decades, we have witnessed an enormously growing amount of data on many IoT (Internet of Things) devices spread across the network. Data on each device is subject to privacy laws. Hence, using data from many different devices for training by machine learning algorithms in the cloud is nearly impossible and poses new challenges. Federated learning tries to cope with this problem by decentralized learning, where each device (client) owns its copy of the machine learning algorithm (most often neural network), trains its private data on this model and sends the results to aggregation on the server. The server collects model parameters from different clients, creates standard models, and returns new parameters to each participating client. This work introduces, compares, and gives an overview of work made using two important techniques for solving optimization problems in federated learning: multi-object learning and multi-task learning. Both techniques were known before, but their application in federated learning is relatively new. They are designed to address important challenges, such as privacy protection, dealing with non-iid (non-independent and identically distributed) data, energy management in wireless networks, decentralized data, etc.
Logically Explained Malware Detection
Peter Anthony, Francesco Giannini, Michelangelo Diligenti, Martin Homola, Marco Gori, Štefan Balogh, Ján Mojžiš
Abstract: The rapid evolution of attack techniques in cybersecurity presents an ongoing challenge for effective malware detection. Traditional signature-based approaches struggle to keep up with the growing volume and diversity of malware. Machine learning offers a promising alternative, but faces limitations in generalizing to unseen threats and providing transparent, human-understandable explanations for malware classifications. In security-critical contexts, where trust and legal compliance are paramount, understanding the rationale behind model decisions is essential. While deep learning models excel in detection accuracy, their ""black-box"" nature impedes explainability. On the other hand, interpretable models often underperform in terms of detection capability. To address this trade-off, we propose the application of Logic Explained Networks (LENs), an emerging class of interpretable neural networks that provide explanations using First-Order Logic (FOL) rules. This study extends the use of LENs to the challenging domain of malware detection, leveraging the large-scale EMBER dataset. Our experimental results demonstrate that LENs offer superior robustness compared to traditional interpretable methods, while achieving performance levels comparable to black-box models. Additionally, we introduce a formalized a version of LENs that generates logic-based explanations with enhanced fidelity to the model's predictions.
Problems of algebraic surgery over ball complexes
Serhii Dylda, Tibor Macko
Abstract: Generalization of algebraic surgery from simplicial complexes to ball complexes is not straightforward, because one runs into a problem of locally finiteness of (co-)products in a categories of ball complexes. In particular, mapping spaces and loop spaces. This presentation is aimed at giving an overview of problems that we have encountered while working on generalization of proof of ""pi-pi theorem"" of algebraic surgery, as welll as attempted/proposed ideas and solution to overcome those problems.
Quantitative analysis of NO2 generated in Atmospheric Pressure Plasma Jet using Ion Mobility Spectrometry
Emanuel Maťaš, Michal Neogrády, Ladislav Moravský, Štefan Matejčík
Abstract: Reactive oxygen or nitrogen species (RONS) like as O3, OH radicals, NO2, H2O2 and other can be generated by Ar plasma jet. There are several standard methods for diagnosing them, such as FTIR spectroscopy or chemical gas sensors. In this work, we show a new possibility NOX detection using ion mobility spectrometry (IMS). The NO2 concentration was determined by NO2- current signal. The detection limit was set at 200 parts per billion (ppb).
Rekonštrukcia neúplných mračien bodov pomocou vonkajšo-vnútorného zmršťovacieho obaľovania varietnými plochami
Pavel Chalmovianský, Róbert Bohdal
Abstract: Nekompletné 3D dáta sú bežným problémom spôsobeným neúplným alebo zatieneným skenovaním. Automatizované procesy, ako je Lagrangiánske zmršťovacie obaľovanie, je možné použiť pre vyplnenie medzier v mračnách bodov. Riešime tzv. problém zastavenia tohto obaľovania zavedením vnútorných plôch, ktoré sa vyvíjajú smerom von, čím dopĺňajú vnútorné sťahovanie vonkajšej plochy. Ukazujeme nutné podmienky pre rovnovážne riešenia a demonštrujeme ich numerickými experimentmi na krivkách a plochách. Naša metóda efektívne dopĺňa medzery v 3D datasetoch a ponúka spoľahlivé riešenie pre rekonštrukciu nekompletných mračien bodov.
Relations between K-theory and Cobordisms
Marián Poturnay
Abstract: Both Topological K-theory and Cobordisms form a generalized cohomology theory. A natural question is to ask what are the relations between these two. Using tools like Bott periodicity and generalized characteristic classes we establish some maps between these two cohomology theories. These maps then allow us to find an isomorphism between K-theory and slightly modified version of cobordisms.
Statistical analysis of microgap vacuum breakdown mechanisms for palladium electrodes in pulsed electric fields
Mária Maťašová, Matej Klas, Peter Čermák, Leonid Satrapinskyy, Štefan Matejčík
Abstract: Nowadays, the non-thermal plasma (NTP) has been used for decades in many fields such as industrial, biological, medical, or environmental. For the uniform treatment of the samples, the homogeneity of plasma is necessary. One of the ways how to ensure plasma homogeneity is its generation in micrometric electrode distances. In such gaps, the high electric field intensities can be achieved ⁓ 10^8 V/m and field emission phenomena occur. Therefore, the main objective of this study is the characterization of field emission current generated in microgap breakdown system between palladium electrodes. However, in high pressures, the fast ionization processes occur that disable the field emission current measurement. For this reason, our measurements are provided in vacuum ⁓ 10^-4 Pa. In this work, the microdischarges were generated in DC pulsed regime with the electric field ramp speed 10^7 kV/s. The electrodes were set in the sphere-plane configuration with a gap size of 1-10 μm. The conditioning of electrodes was performed and saturated stage of the process representing the stage with the stable value of the breakdown voltage in average was presented. Based on the Fowler-Norheim theory, the important parameters for microdischarge characterization can be calculated such as field emission current density, emission area and factor of enhancement of electric field β. In this study, we focused on the statistical analysis for investigation of the relationship of the sequence of successful consecutive breakdowns, for samples over 10,000 sequences, measured in the saturated stage. The memory effect between two consecutive breakdowns was observed.
Siamese color graphs
Nina Hronkovičová, Martin Mačaj
Abstract: The spread of a graph consists of a set of same-size disjoint cliques that span the entire graph. A Siamese color graph of order $q$ is an edge decomposition of the complete graph on $q^3+q^2+q+1$ vertices into $q+1$ strongly regular subgraphs, all of which share the same spread consisting of $q^2 + 1$ cliques of size $q + 1$ and are otherwise edge-disjoint.
In 2003, Kharabani and Thorabi showed that for any prime power $q$ there exists a Siamese graph of order $q$. Later, in 2005, Klin, Reichard, and Woldar further showed that if a Siamese color graph consists solely of geometric strongly regular graphs --- the strongly regular graphs that are point graphs of generalized quadrangles --- then the set of all cliques form a Steiner system with parameters $(2,q+1,q^3+q^2+q+1)$. They used this observation to classify Siamese color graphs of order $2$ and discovered hundreds of geometric Siamese color graph order $3$.
In 2023, we have completed a full classification of geometric Siamese color graphs of orders $3$. Building on this groundwork, our current research explores Siamese color graphs further. Using computer-aided exhaustive search techniques, we extend our investigation to pseudo-geometric Siamese color graphs of order $3$, achieving a comprehensive classification of all Siamese color graphs of this order.
Spectral Analysis of Prominent CaF Molecular Bands in LIBS Spectra of CaF_2 matrix
Shweta Soni, Matej Veis, Pavel Veis
Abstract: This study investigates the spectral characteristics of calcium fluoride (CaF) molecular bands observed in LIBS spectroscopy of in a CaF₂ pellet sample prepared by standard analytical tablets high pressure preparation method from the powder produced from CaF₂ optical windows. Detailed Broadband emission analysis provides insights into plasma cooling dynamics and the timescales of molecular formation within the laser-induced plasma. In the observed LIBS spectra measured at atmospheric pressure under air (delay time 500 ns, gate time 500 ns), both atonic lines (Ca I and Ca II) as well as CaF molecular bands of A-X and B-X transition of for 3 different sequence Δv = 0, ±1. Among these, the vibrational transitions A^2Π_1/2 − X^2Σ^+ (ν_00) at 606.3 nm and B^2Σ^+ − X^2Σ^+ (ν_00) at 531.0 nm are particularly prominent, indicating strong transitions that can be used for further re-excitation using Laser-Induced Fluorescence (LIF). The intensity variations of these transitions, linked to specific molecular processes, provide insights into CaF formation mechanisms within the plasma plume and underscore the potential of LIBS in detailed characterization of fluoride materials.
Acknowledgement: The authors acknowledge the Scientific Grant Agency of the Slovak Republic (contract number VEGA 1/0803/21) and the Slovak Research and Development Agency (APVV-20-0111, APVV-22-0548, and APVV-23-0281), Comenius University (UK/3139/2024) and OeAD/SAIA for support under Aktion Austria - Slowakei (2024-05-15-003)."
Study of Pendant Drop in Electric Field Generated by Electron Beam-Irradiated Hydroxyapatite
Veronika Hidaši Turiničová, Martin Moško, Boris Gajdušek, Juraj Feilhauer, Syed A. M. Tofail, Maroš Gregor
Abstract: Electrostatic interactions play an important role in adsorption of proteins or bacteria on biomaterials. Thus, modification of surface potential can be used to regulate their adsorption. However, a proper understanding of this phenomenon is still lacking. Here we propose a method that can be used to evaluate the effect of modified surface potential on liquids and solutions. It is based on a droplet hanging from a needle (so called Pendant Drop) at a known distance from the modified surface. As the drop approaches the charged biomaterial surface, the electric field generated by the surface charge increases. Free charges or inner dipoles naturally present in the droplet are displaced or oriented, respectively, in the direction of the field. As a result, the upper droplet hemisphere is repelled from the charged surface, lower hemisphere is attracted to it, thus, the drop elongates. In sufficiently high homogeneous or non-homogeneous electric filed, the drop can eventually be torn into two equal subdroplets or detached from the needle, respectively. Moreover, thanks to the modified Laplace equation we can use the modified shape to calculate the value of the charge trapped in the dielectric sample. We demonstrate this measurement using dielectric biomaterial named hydroxyapatite. The pellets of hydroxyapatite were irradiated by electron beam of 30 kV and with electron dosages between 0.01μC/cm2 and 1.0μC/cm2. Behavior of deionized water droplet in proximity of the charged hydroxyapatite was studied and compared with convential coulombmetric measurements. Furthermore, the influence of the sample holder material and geometry was examined.
Study plasma-chemical processes of pentane in atmospheric pressure corona discharge by ion mobility spectrometry-mass spectrometry (IMS-MS)
Priyanka Kumari, Vahideh ILbeigi, Stefan Matejcik
Abstract: The plasma-chemical processes initiated by a Corona Discharge in a pentane-air mixture at atmospheric pressure were studied using Ion Mobility Spectrometry (IMS) and IMS combined with mass spectrometry (IMS-MS). In each experiment, a mixture of air-pentane was directly injected into an external Corona Discharge Reactor (CDR), where it was subjected to the chemical processes initiated by the discharge. Using the designed CDR, it was possible to control the parameters influencing the reactions in the discharge including discharge current, gas composition (air, N2, and O2), pentane concentration, humidity levels, and polarity of the discharge. The neutral products from CDR were analysed by the IMS and IMS-MS techniques and DFT calculations were used to identify the structures of the produced molecules. The main products of pentane produced in the CDR were 5- hydroxy-2- pentanone, 3-pentanone, and 2-pentanone. The product formation changed with humidity as 5- hydroxy-2- pentanone main product at low humidity (0.1%) whereas ketones (3-pentanone and 2-pentanone) produced more at higher humidity (32 %). The method was also used for quantitative determination of pentane, and it was found that the limit of detection (LOD) and the linear range for detection of pentane by the designed system are 0.5 ppmv, 0.5-4.5 ppmv, respectively, and the relative standard deviation (RSD) was approximately 11.0% at the lowest concentration of pentane.
Task Prompt Vectors: Effective Initialization through Multi-Task Soft-Prompt Transfer
Robert Belanec, Simon Osterman, Ivan Srba, Maria Bielikova
Poster
Abstract: Prompt tuning is an efficient solution for training large language models (LLMs). However, current soft-prompt-based methods often sacrifice multi-task modularity, requiring the training process to be fully or partially repeated for each newly added task. While recent work on task vectors applied arithmetic operations on full model weights to achieve the desired multi-task performance, a similar approach for soft-prompts is still missing. To this end, we introduce Task Prompt Vectors, created by element-wise difference between weights of tuned soft-prompts and their random initialization. Experimental results on 12 NLU datasets show that task prompt vectors can be used in low-resource settings to effectively initialize prompt tuning on similar tasks. In addition, we show that task prompt vectors are independent of the random initialization of prompt tuning on 2 different language model architectures. This allows prompt arithmetics with the pre-trained vectors from different tasks. In this way, we provide a competitive alternative to state-of-the-art baselines by arithmetic addition of task prompt vectors from multiple tasks.
Towards CATS: A Modular ABox Abduction Solver Based on Black-Box Architecture
Jakub Kloc, Martin Homola, Júlia Pukancová
Poster
Abstract: CATS, standing for Comenius Abduction Team Solver, is an ABox abduction solver for description logics that allows the users to choose among multiple abduction algorithms. The solver integrates the JFact reasoner as a black-box, thanks to which it handles expressivity up to SROIQ (OWL 2). It handles inputs and outputs via OWL API and offers a command line, GUI, and API interface.
Účinky plazmou aktivovaného PBS na ľudské bunky doplnené o in vitro testovanie dermálnej dráždivosti na modeli pokožky
Darina Kužmová, Helena Gbelcová, Silvia Letašiová, Zdenko Machala
Abstract: Plazmová medicína predstavuje jednu z potenciálnych inovatívnych stratégií v boji proti rakovine. Hlavnú úlohu v plazmovej terapii zohrávajú reaktívne formy kyslíka a dusíka (RONS) generované v plazme a plazmou aktivovaných kvapalinách. Táto štúdia sa zameriava na skúmanie účinkov plazmou aktivovaného fosfátového tlmivého roztoku (PAPBS) na ľudské bunky. Vplyv PAPBS na životaschopnosť buniek je monitorovaný s ohľadom na koncentrácie RONS v PAPBS, a taktiež aj na trvanie jeho aplikácie na bunky, s cieľom určenia optimálnej „dávky“ plazmy s potenciálnym selektívnym antitumorovým účinkom. Keďže niektoré typy nádorov môžu spôsobovať ulcerácie kože, je možné, že pri aplikácii PAPBS na takéto nádory dôjde aj ku kontaktu so zdravou pokožkou v okolí. Preto je dôležité otestovať, bezpečnosť PAPBS, aby sa predišlo dermálnej toxicite alebo podráždeniu. Na testovanie potenciálnej citlivosti pokožky na PAPBS bol použitý EpiDermTM, patentovaný model pokožky vyvinutý spoločnosťou MatTek.
Understoichiometric ZrB2-x films prepared by High Power Impulse Magnetron Sputtering (HiPIMS)
Viktor Šroba, Tomáš Fiantok , Marin Truchlý, Tomáš Roch, Branislav Grančič, Katarína Viskupová, Leonid Satrapinskyy, Peter Švec, Jr., Štefan Nagy, Vitalii Izai, Peter Kúš, Marián Mikula
Abstract: Diborides of transition metals (TM) from IVB group are materials with excellent mechanical properties. Thin films based on diborides are often prepared using magnetron sputtering from a compound target leading to growth of overstoichiometric TMB2+x films (B/TM > 2). These films exhibit nanocomposite structure comprised of stoichiometric few nanometers wide TMB2 filaments with hexagonal crystal structure that are surrounded by excess boron in form of an amorphous tissue phase. To improve the mechanical properties and oxidation resistance at elevated temperatures, control over volume fraction of boron tissue is needed. Technology called high power impulse magnetron sputtering (HiPIMS) uses short high-power pulses to ionize significant portion of sputtered material, that can be attracted towards the growing film. This way, understoichiometric TMB2-x films with higher concentration of TM can be prepared.
Vplyv rozvoja rôznych typov matematického myslenia na schopnosti žiakov s medzipredmetovým presahom
Dušan Daniel
Abstract: Predmetom nášho výskumu je skúmanie vzťahu medzi geometrickým a algebraickým myslením. V diplomovej práci sme sa venovali tvorbe úloh z vybraných oblastí stredoškolskej geometrie so zameraním na podporu argumentačných zručností. Na základe realizovaných analýz sme zistili, že implementácia takýchto úloh môže pozitívne prispieť k rozvoju geometrického myslenia žiakov. V dizertačnej práci sa zameriavame na identifikáciu vzťahu medzi úrovňami rozvinutosti algebraického a geometrického myslenia a schopnosťou žiakov riešiť vybrané informatické úlohy. Na základe zistení zostavíme súbor vyladených úloh, zameraných na obsahový komponent ""Závislosti, vzťahy a práca s údajmi"", ktoré budú určené pre tretí cyklus. Tento komponent predstavuje prierezovú tému pre vyučovanie matematiky a informatiky. V rámci našej práce budeme tieto úlohy opakovane testovať a pravidelne sledovať a analyzovať stratégie riešenia žiakov s cieľom zhodnotiť ich posun v oblasti algebraického a geometrického myslenia a pokrok v riešení informatických úloh. Práca by mala priniesť teoretický model, ktorý popíše, ako jednotlivé zložky algebraického a geometrického myslenia ovplyvňujú schopnosť žiakov riešiť informatické úlohy, a zároveň naznačí, akým spôsobom možno tieto schopnosti rozvíjať prostredníctvom vyučovania prierezových tém medzi matematikou a informatikou.
Vzdelávacia hra Fyzikálny brainbox
Lívia Sobinovská, Tünde Kozánek Kiss
Poster
Abstract: Poster je ukážkou časti záverečnej práce študentky učiteľstva fyziky prezentovaný na festivale Science on Stage vo Fínsku v auguste 2024. Fyzikálny brainbox je vzdelávacia hra, ktorá vznikla ako modifikácia komerčnej hry Brainbox. Je predovšetkým zameraná na témy z oblasti fyziky a je využiteľná vo vyučovaní fyziky na základnej škole. Hru Fyzikálny brainbox môže učiteľ zaradiť do vyučovania ako opakovanie, ale aj na rozvíjanie a získavanie faktických poznatkov. Rovnako ako v hre Brainbox, aj v nami vytvorenej verzii tejto hry sa nachádza na prednej strane kartičky ilustrácia a na zadnej strane 6 otázok. K hre Fyzikálny brainbox sme zatiaľ vytvorili 10 hracích kartičiek z kategórie Sily a tlak v kvapalinách. Hru Fyzikálny brainbox plánujeme rozšíriť o ďalšie kategórie. Témy jednotlivých kategórií budú zhodné s témami z fyziky, ktorým sa venujú žiaci základných škôl.
What Pre-conceptions Do Students Have About Sensors? Preliminary Results
Barbora Stenová, Karolína Miková
Abstract: Education that prepares young people for their future lives should naturally take into account the current epoch and its technological developments. The integration of robotics into everyday life is becoming more and more commonplace. Students thus arrive at school with certain ideas, not always good ones, about how things or robots in their environment work. This subconsciously acquired knowledge, the so-called pre-concepts, has been changing considerably over the last years. Teaching materials, learning objectives and education in general should respond to or take account of this change. However, there is no precise information about these learner pre-concepts. In our dissertation research, we therefore focus on students' pre-concepts about the robots that students come to class with. In this paper, we examine specific pre-concepts about sensors and how students explain their operation. We used mixed methods data collection and analysis, where we analyzed questionnaire data from 48 students aged 11-13. In this paper we present the 4 different pre-concepts identified.

Poster Session 15:00-16:30 - Results of departments and research groups

AI and Bioinformatics-Based Stain Identification System for Historic Paper Conservation and Restoration
Zuzana Černeková, Zuzana Berger Haladová, Andrej Ferko
Abstract: The restoration and conservation of works on paper often involve removing various types of stains from the surface of the paper material. These stains can have different origins—ranging from microbiological (such as fungi and bacteria), and insect droppings, to ink stains and stamps. Accurate recognition and early identification of microbial contamination or infestation are essential for effective preventive measures and optimizing restoration procedures. In this context, an application for stain identification would be valuable during the monitoring of collections, storage areas, or during the conservation and restoration processes. This project focuses on utilizing machine learning and artificial intelligence in image analysis to improve diagnostic accuracy and develop a method for the rapid identification of stains on historical paper. This method would be applied within the monitoring of collections and conservation processes.
Ako dostať súčasnú vedu k učiteľom fyziky
Tünde Kozánek Kiss
Abstract: Oddelenie didaktiky fyziky Katedry didaktiky matematiky, fyziky a informatiky každoročne organizuje odbornú konferenciu Šoltésove dni pre učiteľov fyziky základných a stredných škôl. Konferencia sa uskutočňuje formou dielní, ktoré sú doplnené odbornými prednáškami. Dielne väčšinou zabezpečujú členovia Oddelenia didaktiky fyziky. Čo sa týka odborných prednášok, časť zabezpečujú členovia rôznych katedier našej fakulty. Vďaka nim sme učiteľom priblížili napríklad fyziku plazmy, alebo to, ako vznikajú zemetrasenia. Taktiež si učitelia mohli vypočuť prednášku o radóne a jeho účinkoch na ľudské zdravie alebo o elektromobiloch. Ďalšími témami boli napríklad neutrína, vesmírny ďalekohľad Jamesa Webba alebo svetelné znečistenie. Vďaka týmto (a ďalším) prednáškam od našich kolegov z fakulty sa snažíme učiteľom z praxe, a taktiež budúcim učiteľom fyziky, priblížiť súčasnú vedu.
Ako presné môže byť numerické modelovanie seizmických vĺn?
Jaroslav Valovčan, Jozef Kristek, Martin Gális, Miriam Kristeková
Abstract: Numerické modelovanie šírenia seizmických vĺn metódou konečných diferencií sa rozvíja od polovice 60-tych rokov 20. storočia. Prirodzene sa seizmológovia usilovali o modelovanie vĺn v reologicky i geometricky realistických modeloch vnútra Zeme. Za uplynulých 60 rokov vznikli stovky konečno-diferenčných schém rôznej presnosti a rôznej úrovne konzistencie s fyzikou útlmu, geometrie materiálových rozhraní a topografie voľného povrchu. Z dôvodov, ktoré nie sú zrejmé, seizmológovia pozabudli na principiálne súvislosti medzi nehomogenitou prostredia, vzorkovaním v priestorovej oblasti a vlnovo-číselným spektrom vlnového poľa. Analýza týchto súvislostí nás doviedla k zásadným zisteniam o limitoch presnosti numerického modelovania. Výsledky tvoria tri články v Bulletin of the Seismological Society of America.
Ako sa skúmajú historické zemetrasenia: zemetrasenia v okolí Dobrej Vody 1906 a 1930
Róbert Kysel
Poster
Abstract: Územie Slovenska patrí medzi oblasti s miernou seizmickou aktivitou. Existujú však archeologické dôkazy a písomné pramene o zemetraseniach v minulosti, ktoré spôsobili škody na budovách a straty na ľudských životoch (napr. stredoslovenské zemetrasenie v roku 1443, komárňanské zemetrasenia v rokoch 1763, 1783, 1806, 1822 a 1851, žilinské zemetrasenia v rokoch 1600, 1613 a 1858, východoslovenské zemetrasenia v rokoch 1885, 1890, 1893, 1914, 1932 a 1941). Severná časť Malých Karpát patrí medzi seizmicky najaktívnejšie oblasti Slovenska s historickými zemetraseniami v rokoch 1906 (najsilnejšie zemetrasenie na území Slovenska v 20. storočí). a 1930. V príspevku vysvetlíme metodologický postup pri skúmaní historických zemetrasení, ako aj to, prečo je pre výskum zemetrasení potrebný archívny výskum a aké informácie môže o zemetraseniach priniesť dobová tlač.
Alignment-free de novo structural variations from pangenomes
Luca Denti, Paola Bonizzoni, Brona Brejova, Fereydoun Hormozdiari, Tomas Vinar
Abstract: Structural variations are medium to large size chromosomal rearrangements that play a key role in many genetic diseases and evolution. State-of-the-art approaches for structural variations discovery mainly depend on expensive de novo genome assembly or error-prone long read alignment to linear reference genome. This makes structural variations characterization in complex regions of the genome (e.g., low complexity regions and large segmental duplications) not practical nor reliable. We propose the first alignment-free approach for characterizing structural variations that are novel w.r.t. a set of genomes, i.e., a pangenome. By cleverly combining string-based and graph-based representations of a pangenome, we characterize de novo structural variations with respect to all individuals of the population under investigation. Our approach pinpoints breakpoints using specific strings, anchors and clusters them using a k-mer based sketch of the input graph, and then locally reanalyze the corresponding subgraphs by applying a de novo assembly inspired technique. As proven by an extensive evaluation on simulated data, our approach is able to achieve the accuracy of methods based on de novo assembly without being negatively affected by misalignments that commonly occur when aligning long reads over complex regions.
AMOS - unikátna celosvetová sieť vlastných kamier
J. Tóth, L. Kornoš, T. Paulech, P. Zigo, P. Matlovič, M. Baláž, D. Kalmančok a J. Šimon
Abstract: AMOS - All-Sky Meteor Orbit System je originálny astronomický prístroj vyvinutý na FMFI a AGO Modra, ktorý sa rozšíril na Slovensko a svetové prestížne astronomické pracoviská, respektíve na miesta s "dokonalou" nočnou oblohou. Spolupráca s kolegami z fakulty, priemyslu, našimi absolventami v zahraničí ako aj zahraničnými partnerskými univerzitami a inštitúciami je nevyhnutná pri zostavení a prevádzkovaní tejto jedinečnej celosvetovej siete kamier zameranej na výskum meteorov a meteoritov ako aj kozmického odpadu.
Analysing edge-on galaxies with deep learning
Žofia Chrobáková, Viera Krešňáková, Peter Butka, Júlia Gazdová
Abstract: The advent of large astronomical surveys, such as Euclid, will offer unprecedented insights into the statistical properties of galaxies. However, the large amounts of data that will be generated by these surveys call for the application of machine learning methods. For this purpose, we trained the YOLOv5 algorithm to accurately detect spiral, edge-on galaxies in astronomical images and the SCSS-Net neural network to generate segmentation masks, so that the detected galaxies can be used for any further analysis. This algorithm was applied to current astronomical images; however, its real power lies in its applicability to data from future surveys, where it can lead to new discoveries. We will also present one of our future goals, which is the study of the galactic warps: a well-known distortion of the galactic discs occurring in most spiral galaxies, including the Milky Way. Though hundreds of warped galaxies are known, the mechanism behind warp formation remains unclear. Our work will offer a comprehensive statistical analysis that will explore the relationship between different warps, their dependence on environmental factors, and their role in galactic evolution.
Analysis of the shear stress/shear rate relationship of whole blood treated with dendritic nanoparticles at various concentrations
Šimon Šutý, Patrick Mydla, Iveta Waczulíková
Abstract: For tumor-targeted drug delivery, intravenously administered nanoparticles circulate in the bloodstream and must be small enough to escape circulation through tumor microvasculature. However, events such as a premature release of the drug from nanovehicle still may occur, which results in unintended interactions with blood components. Blood compatibility and safety of any nanoformulation for medical use must therefore be carefully evaluated before. In our study we used a first (DD1) and a second generation dendrimer (DD2) at concentrations of 0, 2, 10 and 30 μM which had already been tested in cell studies. Since blood is considered a non-newtonian fluid, we focused on the flow behaviour and shear stress/shear rate relationship upon interaction of blood with DD1 and DD2. We used a rotational modular rheometer in the range of shear rate of 1 to 1000 s-1 to mimic circulation in various in vivo conditions. We observed an increase in shear stress at all concentrations of both dendrimers. These effects are correlated with hemocoagulation parameters presumably without physiological significance. However, higher doses may pose increased risk under in vivo conditions ranging from problems with perfusion in body tissues to thrombosis and embolism. Funded by the EU NextGenerationEU through the Recovery and Resilience Plan for Slovakia under the project No. 09I03-03-V04- 00375.
Antimicrobial Activity of SDBD Plasma Treatment on Planktonic and Biofilm Forms of Escherichia coli
Oleksandr Galmiz, Bernard Gitura Kimani, Zdenko Machala
Poster
Abstract: By using a water solution as a discharge electrode, it is possible to combine the basic features of both water discharges and surface dielectric barrier discharges (SDBD). The benefit of such a combination lies in obtaining highly oxidative plasma in contact with liquid without problems of electrode erosion. The SDBD is generated from the contact line between the liquid electrode, air, and dielectric material (tube). Such combined systems were used for plasma activations of hollow object surfaces, which performed splendidly in achieving uniform plasma treatment of the inner surface of fine tubes. This study investigated several approaches to plasma interaction with bacteria. The first scenario is the effect of plasma on the liquid contaminated with planktonic bacteria E. coli (CCM 3954). Various power inputs and treatment times were investigated. The second approach is to determine the effect of plasma treatment on E. coli 48h biofilms deposited on the inner surface of polymeric tubes. In the third approach, the plasma pre-treatment of polytetrafluoroethylene (PTFE) tubes before biofilm formation was performed to investigate its ability to change the microbial surface adhesion and therefore influence biofilm maturation by changing the surface properties of the PTFE tubes. Our results show that plasma treatment reduces microbial water contamination, impedes the biofilm formation process, and decreases the biomass of the mature biofilm. This work was Funded by the EU NextGenerationEU through the Recovery and Resilience Plan for Slovakia under the project No. 09I03-03-V04-00094 and by Marie S. Curie Action Postdoctoral Fellowship under Horizon Europe with grant agreement number 101066764.
Antiyeast Activity of Plasma Activated Water Irradiated With UV A and Combined With Natural Phenolics on Planktonic Wickerhamomyces anomalus
Bernard Gitura Kimani, Ramin Mehrabifard, Oleksandr Galmiz, and Zdenko Machala
Poster
Abstract: Plasma-activated water (PAW) has emerged as a potent antimicrobial tool with the ability to control microbial growth [1]. PAW is produced by treating water with non-thermal plasma, introducing various reactive oxygen and nitrogen species (RONS) into the water thus making the treated water possess antimicrobial properties [1]. Phenolic compounds, predominantly found in plants, have demonstrated significant potential as anti-yeast agents by disrupting cellular processes [2]. The combination of phenolics and PAW presents a promising anti-yeast strategy with potential application in medical, food safety, agriculture, and water treatment settings [3]. The synergistic effect of combining phenolics with PAW is attributed to their complementary mechanisms of action, which may include a wider spectrum of antimicrobial activity, reduced antimicrobial resistance, increased oxidative stress and enhanced membrane disruption [4]. This study involved the use of 1 kHz transient spark (TS) discharge PAW generated in the presence of UV-A and combined with cinnamic acid, vanillin, gallic acid and p-coumaric acid separately to create phenolics concentration of 2 or 1 mg/mL, and incubated with 106 CFU/mL of yeast W. anomalus SZMC 8061Mo for 24 hours at 30 oC. In addition, 106 CFU/mL of W. anomalus in sterile tap water was directly treated with TS for 10 min in the presence of UV-A and incubated with each of the four natural compounds under similar conditions. The efficacy of the PAW/UV-A radiation combined with natural phenolics was evaluated by assessing the growth extent after 24 hours’ incubation through agar plated colony counts and compared to the untreated control samples. The findings of this research indicate that the combination of PAW/UV-A with natural phenolics represents a viable anti-yeast strategy.
Acknowledgment: This work was supported by Funded by the EU NextGenerationEU through the Recovery and Resilience Plan for Slovakia under the project No. 09I03-03-V03-00033 and Slovak Research and Development Agency APVV-22-0247 and Marie S. Curie Action Postdoctoral Fellowship under Horizon Europe with grant agreement number 101066764 for O. Galmiz.
References [1]. Rahman M. et al., Int. J. Environ. Res. Public Health. 19, 6630, (2022). [2]. Takó M. et al., Antioxidants. 9, 165, (2020). [3]. Xu H. et al., J. Chem. Eng. 452, 139596, (2023). [4]. Prasad K. et al., Molecules. 28, 7481, (2022).
At Least a Rainbow Flag: Queer Students' Perceptions of University Responses to Queerphobic Terrorist Attack
Mária Čujdíková, Xenia Daniela Poslon, Viktor Kocur
Abstract: This paper explores how queer students perceived the higher education institutions' responses to a deadly queerphobic terrorist attack which took place on October 12, 2022 in Bratislava, Slovakia. Drawing on data collected after the attack from 34 college students from Slovakia, the study employs thematic analysis to uncover key themes. Symbolic gestures, such as displaying the pride flag, were described by participants as important steps toward inclusion, but the study reveals the need for institutions to move beyond symbolism. The participants emphasize the importance of inclusive policies, training programs, counseling resources, and real systemic changes to create truly safe and inclusive spaces. The paper underscores concerns about institutional reluctance to implement such measures, despite research supporting their positive effects.
COLOSSE project
Marián Mikula
Abstract: Horizon Europe project at CENAM.
Cyclic connectivity of cages
Robert Lukoťka, Edita Máčajová
Abstract: We shall investigate the cyclic connectivity of the cage graphs, where a \emph{$(k, g)$-cage} is a smallest $k$-regular graph with girth $g$. A graph $G$ is cyclically \emph{$c$-edge-connected} if one needs to remove at least $c$ edges to disconnect $G$ into at least two cyclic components. Cyclic connectivity proved to be a profitable structural property of regular graphs, especially cubic ones. The cyclic connectivity of cages has not been studied previously. We prove that if a $(k, g)$-cage $G$ has at most $2M(k, g) - O(g^2)$ vertices, where $M(k, g)$ is the Moore bound, then $G$ is cyclically $(k - 2)g$-edge-connected and every cycle-separating $(k - 2)g$-edge-cut in $G$ separates a cycle of length $g$. In particular, this is true for unknown cages with $(k, g) \in \{(3, 13), (3, 14), (3, 15), (4, 9), (4, 10)$, $(4, 11),$ $(5, 7), (5, 9), (5, 10), (5, 11), (6, 7), (9, 7)\}$ and also the potential missing Moore graph with degree $57$ and diameter $2$. Also, we conjecture that all cages have these properties.
Časticová fyzika na ATLASe
Róbert Astaloš, Pavol Bartoš, Michal Dubovský, Adriana Dohnalová, Barbora Eckerová, Lucia Keszeghová, Matúš Kosťov, Ivan Sýkora, Daniel Suchý, Stanislav Tokár, Tibor Ženiš
Abstract: Poster prezentuje náš príspevok k výskumu venujúcemu sa časticovej fyzike na experimente ATLAS v CERN-e. Naša skupina skúma vlastnosti najťažšej elementárnej častice - top kvarku, no venuje sa i skúmaniu oblasti, v ktorej sa z kvarkov a gluónov vytvárajú hadróny. Zároveň sa vo veľkej miere venujeme kvalite zaznamenávaných dát v sub-detektore TileCal, ktorého úlohou je meranie energie hadrónov.
Delta-T Noise, a New Tool for Detecting Temperature Fields in Superconducting Films
Elena Zhitlukhina, Paul Seidel
Poster
Abstract: The fundamental issue generated by finite temperatures is spontaneous and uncorrelated oscillations of electric currents. Recently, a new (previously overlooked) noise contribution caused by the temperature difference in the conductive device and referred to as delta-T noise was found. In the work, we apply this idea to a normal-metal – superconductor contact and show that measurements of the non-equilibrium delta-T noise, along with the steady-state characteristics, can be used for revealing dynamical finite-frequency processes inherent in the samples.
Designing a cell and deeply cooled amplifier for quantum measurements
Oleh Turutanov, Alexey Korolev, Martin Baránek, Samuel Kern, Valery Lyakhno, Pavol Neilinger, Miroslav Grajcar
Abstract: Single-photon detectors are key elements of prospective quantum techniques. We proposed a scheme for a single microwave photon counter based on a superconducting flux qubit. To make weak continuous measurements of the qubit, it should be isolation from electromagnetic environment and thermalized down to 10-20 mK. Here we report a design of the measuring cell with emphasis on cooled ultra-low dissipation HEMT amplifier having low back action. The cell is surrounded by a three-layer superconductor and ferromagnetic shield with calculated attenuation of 200 dB (dc fields) and 80 dB (low-frequency ac fields). The sample holder is a dielectric sapphire rod to exclude Nyquist noise which otherwise decoherences the qubit. An rf SQUID to sense a small change (~0.3 magnetic flux quantum) in qubit’s magnetic flux operates in adiabatic non-hysteretic regime at 450 MHz pumping and low rf driving amplitude. It is placed on the same chip with the qubit. The SQUID tank circuit voltage is put to a cold HEMT-based pre-amplifier, which works in unsaturated regime and dissipates 1-2 microwatt per HEMT. The brightness temperature of the HEMT input in this regime is close to its physical temperature, so its Planck irradiation and hence back action are negligible at the working frequency. The amplifier design is discussed.
Differences in Approaches to Reasoning and Proving in Textbooks from Four EU Regions on the Topic of Thales' Theorem
Mária Slavíčková, Kinga Szűcs
Poster
Abstract: The presentation will showcase the results of an analysis of 12 textbooks from the Czech Republic, Hungary, Slovakia, and Thuringia (i.e., three from each region) on the topic of Thales' Theorem, with a focus on reasoning and proving activities in both the explanatory and practice sections of the textbooks. The chosen approaches for textbook analysis highlight the unique position of Hungarian textbooks and reveal common features regarding the types of activities in pairs of textbooks.
Discrete dynamical systems with random impulses
Jozef Kováč, Ján Veselý, Katarína Janková
Poster
Abstract: In discrete dynamical systems generated by a continuous mapping f of a real compact interval into itself, we study the effect of impulses at random times. By impulses, we mean that at random times the impulse function g is used instead of the function f . We show the conditions that ensure the stochastic stability of the system.
Environmentálne a biomedicínske aplikácie plazmy
Mário Janda, Zdenko Machala, Karol Hensel
Abstract: Žijeme vo svete ktorý čelí mnohým environmentálnym výzvam týkajúcich sa znečistenie vody, vzduchu a globálneho otepľovania zapríčineného ľudskou činnosťou zvyšovaním koncentrácie CO2 (a iných plynov) v atmosfére. Z tohto dôvodu je nutné zaoberať sa hľadaním a skúmaním nových technológií ktoré by boli použiteľné pri riešení týchto problémov, alebo by sa týmito technológiami nahradili súčasne používané technológie za účelom zníženia environmentálnych dopadov pri tvorbe rôznych chemikálií. Medzi takéto nové environmentálne technológie patria aj procesy využívajúce elektrické výboje a nerovnovážnu plazmu (NP). Týmto environmentálnym plazmovým technológiám sa venuje aj náš tím. Okrem toho sa venujeme aj aplikáciám NP v medicíne a poľnohospodárstve. Aby sme týmto technológiám mohli lepšie porozumieť a ďalej ich vyvíjať, potrebujeme základné znalosti o plazme a o plazmochémii. Preto sa venujeme aj základnému výskumu NP použitím pokročilých diagnostických metód a chemickým kinetickým modelovaním. Výskum je podporený grantom “EÚ NextGenerationEU: Plán obnovy a odolnosti SR č. 09I03-03-V03-00033“.
Extremal structures in Discrete Mathematics
Robert Jajcay
Poster
Abstract: In our poster presentation, we aim to introduce the main concepts and goals of our APVV grant APVV-23-0076
FERTILE: Umelecká edukačná robotika na podporu informatického myslenia v kontexte kombinovaného vzdelávania
Karolína Miková, Kubincová Zuzana, Budinská Lucia, Krcho Jakub, Stenová Barbora
Abstract: Integrácia edukačnej robotiky (ER) do vzdelávacej praxe je spojená s rozvojom digitálnych zručností a informatického myslenia (IM), čo pozitívne ovplyvňuje osobnostný rozvoj študentov. Keďže ER bola hlboko spojená s prezenčným vyučovaním v triede a interakciou tvárou v tvár, pandémia Covid-19, ktorá nám nanútila online učenie náhle prerušila implementáciu robotiky na všetkých stupňoch vzdelávania od základného cez stredné až po vysoké školy. Hoci sa objavilo niekoľko iniciatív na vývoj simulátorov ER, ktoré boli nedávno vydané aj ako nákladovo efektívne riešenie v prípade, že robotické technológie nie sú škole dostupné, ich využívanie zostáva zriedkavé. Okrem toho neexistujú žiadne metodiky na podporu pedagógov pri navrhovaní a realizácii kombinovaných (čiastočne prezenčných, čiastočne online) alebo úplne online projektov výučby robotiky. Zdá sa teda, že v postpandemickej ére sa otvárajú možnosti pre digitálnu transformáciu, pričom sa skúma potenciál integrácie ER do zmysluplného kontextu kombinovaného vzdelávania.
Fluid Dynamics of Phase Change
Peter Guba, et al.
Abstract: Solidification of multicomponent alloys has one important feature in common: the appearance of a mushy layer. Mushy layers are multiphase porous mixtures of dendritic solids and solute-rich interstitial fluid. They host complex dynamics of fluid flow, thermal and chemical transport, phase transformations, nonlinear dynamics and pattern formation. We present the results of mathematical modelling of phase transformations and fluid flow in the mushy layers of binary and ternary alloys. Applications of theory and numerical simulations range from the degradation of industrially cast turbine blades for the aerospace and energy industries to the growth of sea ice.
High temperature boride coatings for aerospace industry
Tomáš Roch, Marián Mikula, Martin Truchlý*, Tomáš Fiantok*, Branislav Grančič, Vitalii Izai*, Leonid Satrapinskyy, Pavol Ďurina, Marek Vidiš *, Maroš Gregor, Katarína Viskupová, Viktor Šroba, Marián Haršáni, Ján Bajús, Martin Frkáň, Adam Frkáň, Dušan Volna, Jiří Šilha
Abstract: This contribution presents our two activities under contract with the European Space Agency (ESA) in the frame of Plan for European Cooperartion States (PECS). Both projects are research oriented preparatory type of studies – proof of concepts. In the first project we deal with ""Potential solid lubricant for extreme temperatures based on vanadium boride"". There are currently no suitable solid lubricants for high temperatures in space environmental conditions. The transition metal (TM) alloyed vanadium boride compounds could be promising candidates because of high thermal stability, high hardness, and possible oxidation products with low friction at intermediate and high temperatures. Understoichiometric VBx, and alloying with Mo and W showed promising results. In the second project we proposed homogeneous, dense, non-porous interlayer oxide suitable scale coating as an active barrier preventing the oxygen penetration to protect components exposed to extreme thermal environments. We investigated chemical and structural properties of Zr-Si-Ta-B based coatings produced by high-power impulse magnetron sputtering. The thermal stability testing and pulsed laser ablation and oxyacetylene torch validation in laboratory environment have been performed.
Inference of stochastic dynamics from tracked biological data
Katarína Boďová, R. Kollár (FMFI UK), G. Tkačik, B. Cassilas-Pérez, S. Cremer, D. Brückner (ISTA), R. Harpaz, E. Schneidman (Weizmann Inst.)
Poster
Abstract: Recent advances in experimental biological techniques have generated rich datasets, allowing for a deeper understanding of the complex dynamics underlying biological processes. However, unraveling the interactions between these underlying components presents an inverse problem, where the challenge is to separate systematic dynamical patterns from inherent stochasticity and unavoidable measurement errors. We present our recent approach for inferring the laws governing collective animal motion, which incorporates the discrete, stochastic, and internal-state-dependent nature of animal behavior. We demonstrate how this method reveals the individual decisions that lead to collective disease defense in ants. Finally, we introduce a novel approach, based on improvements to the method of Frishman & Ronceray (2020), which addresses a common limitation of standard inference techniques—their reliance on exceedingly large datasets with fine temporal resolution.
Information flow security
Damas Gruska
Poster
Abstract: System security based on the absence of information flows is an important part of the development and maintenance of systems of various nature. Formal methods play an important role in how to define, specify and verify such security. Several formalisms and procedures will be mentioned in the presentations.
Inovačné technológie učenia v príprave budúcich učiteľov matematiky
Peter Vankúš
Abstract: V príspevku predstavíme projekt KEGA s rovnomenným názvom. Projekt sa zameriava na integráciu inovatívnych vzdelávacích technológií vrátane umelej inteligencie (AI), gamifikácie a online vzdelávacích únikových miestností do prípravy budúcich učiteľov matematiky. Cieľom projektu je zistiť, ako môžu tieto špičkové technológie zlepšiť vzdelávacie skúsenosti a výsledky budúcich učiteľov matematiky v rámci ich vysokoškolského štúdia, s konečným cieľom zlepšiť ich pripravenosť efektívne vyučovať matematiku v prostredí základných a stredných škôl.
Kvantový priestor
Juraj Tekel, Samuel Kováčik
Poster
Abstract: Uvažuje sa (už pomerne dlho), že na veľmi malej škále má priestor kvantovú štruktúru. V práci sa venujeme viacerým aspektom tejto hypotézy – jej matematickým základom, simuláciám na počítači či možným satelitným pozorovaniam. V príspevku si zhrnieme základné historické míľniky a argumenty v prospech tejto myšlienky, jej dopad na ostatné fyzikálne teórie a význam v širšom kontexte teoretickej fyziky.
Low-Energy Electron Processes wit Molecules and Molecular Clusters in Gas-Phase
Peter Papp, Dušan Mészáros, Peter Samuel Kovár, Štefan Matejčík
Poster
Abstract: Low-energy electrons play important role in various processes in nature as well as in current industry. Electron induced processes have been identified as crucial for different fields of science related to studies of biological systems, atmospheric physics, astrochemistry and plasmas in interstellar space and solar system objects. The importance of understanding electron–molecule and electron-cluster interactions has been highlighted in recent decades for industrial applications as well, including plasma processing, fusion, nanotechnology and health care. The Electron and Plasma Processes Laboratories (EPPL) at the Comenius University in Bratislava are well recognized in the field of low-energy electron processes on atoms, molecules and clusters in gas-phase. Electron ionization, attachment and the related dissociative processes are measured by the means of mass spectrometry and interpreted with quantum chemical calculations. Most recently Plasma-Enhanced Chemical Vapor Deposition (PECVD) and Focused Electron Beam Induced Deposition (FEBID) precursors were studied using crossed electron/molecular beams techniques and theory.
Mathematical Programming Approaches in Data Envelopment Analysis
Mária Trnovská, Margaréta Halická, Jana Szolgayová, Jakub Hrdina
Poster
Abstract: Data Envelopment Analysis (DEA) is a field of Operations Research that focuses on non-parametric methods for evaluating the performance of decision-making units. Each DEA model is formulated as a mathematical programming problem. Based on the structure and objectives of these models, they can be broadly classified into two main categories: slacks-based and path-based models. DEA models are represented in two forms: the envelopment and the multiplier form, which are connected through a primal-dual relationship. Consequently, the duality theory in mathematical programming provides a valuable framework for the theoretical analysis of DEA models, aiding in their appropriate application and the credible interpretation of results. Moreover, selecting the right optimization algorithm can enhance computational efficiency and improve the practical evaluation of decision-making unit performance. (Poster presentation)
Maths meets practice
Daniela Majerčáková, Gábor Szűcs
Abstract: V našom príspevku predstavíme spoluprácu dvoch matematických katedier Fakulty matematiky, fyziky a informatiky Univerzity Komenského v Bratislave (FMFI UK) s praktickou sférou. Katedra aplikovanej matematiky a štatistiky a Katedra matematickej analýzy a numerickej matematiky aktívne a dlhodobo spolupracujú s niektorými subjektmi finančného, bankového, poistného, analytického, výskumného, podnikateľského a konzultačného sektora, ako aj s rôznymi štátnymi orgánmi a inštitúciami. Pre študentov FMFI UK organizujeme pozvané prednášky, praktické semináre a workshopy, ktoré im počas štúdia umožňujú získať cenné praktické skúsenosti. S cieľom prispieť k lepšiemu prepojeniu akademického vzdelávania s praxou, počas minulých akademických rokov sme sa podieľali aj na príprave niekoľkých doplnkových kurzov, ktoré pre študentov FMFI UK ponúkali niektoré partnerské spoločnosti našich katedier. Tento príspevok plánujeme prezentovať formou konferenčného posteru.
Memory Effect in Pulsed Vacuum Microdischarges
M. Klas, M. Maťašová, P. Čermák, Š. Matejčík
Abstract: The understanding of microgaps breakdown phenomena at controlled laboratory conditions represents an important prerequisite in the development of future micro-plasma technologies. The present work is focused on the observation of the electrical discharge parameters of the pulsed vacuum breakdown between palladium electrodes.A detailed statistical analysis of the discharge parameters, such as the breakdown voltage, the duration and the energy released during a pulse, and minimal power for maintaining the discharge is provided for different gap configurations for a voltage ramp speed of 10^7 kV/s. The analysis with respect to breakdown voltage yielded the observation of the memory effect between subsequent breakdowns.
Minimum-Volume Enclosing Ellipsoids: Computation and Statistical Applications
Radoslav Harman, Lenka Filová, Samuel Rosa, Pál Somogyi
Poster
Abstract: We present an efficient algorithm for computing the minimum-volume enclosing ellipsoid for large datasets, with a particular emphasis on tall datasets—those characterized by a vast number of data points in moderate dimensions. We highlight the algorithm’s motivations, grounded in the theory of optimal experimental design, and showcase its practical applications in robust statistics and anomaly detection. In particular, we demonstrate its utility in identifying potential outliers in large datasets through ellipsoidal trimming techniques.
Modelovanie dôchodkového systému na Slovensku
Igor Melicherčík, Gábor Szűcs
Abstract: Príspevok sa venuje dvom aspektom dôchodkového systému na Slovensku: výplatnej fáze starobného dôchodkového sporenia (v tzv. II. pilieri) a rodičovskému bonusu v starobnom dôchodkovom poistení (v tzv. I. pilieri). Pri výplatnej fáze diskutujeme o rôznych zmysluplných spôsoboch použitia úspor z II. piliera. Okrem štandardnej doživotnej anuity špeciálnu pozornosť venujeme aj poisteniu dlhodobej starostlivosti v prípade nesebestačnosti. Rodičovský bonus v dôchodkovom systéme na Slovensku už do určitej miery dlhodobo existuje. Zvýraznil sa nedávnym zavedením rodičovského dôchodku. V príspevku diskutujeme o miere závislosti dôchodku od individuálnych rozhodnutí týkajúcich sa plodnosti pri rôznych variantoch rodičovského bonusu. Snažíme sa kvantifikovať optimálny stav a navrhnúť spôsob, ako sa k tomuto optimu priblížiť. Príspevok plánujeme prezentovať formou posteru.
Modelovanie záporných úrokových mier
Beáta Stehlíková, Michal Jánoši, Lenka Košútová, Mária Rudolfová, Patrícia Ustaníková
Abstract: Poster prezentuje výsledky dosiahnuté pri riešení grantu "Vega 1/0760/22 Modelovanie záporných úrokových mier", ktoré sa týkajú štatistickej analýzy úrokových mier, oceňovania dlhopisov pomocou parciálnych diferenciálnych rovníc a odhadovania parametrov modelov.
Neutrínový teleskop KM3NeT v stredozemnom mori ako nástroj na skúmanie skrytých kútov nášho vesmíru
Rastislav Dvornický, Eliška Eckerová, Zuzana Bardačová, Fedor Šimkovic
Poster
Abstract: Zdroje a mechanizmy urýchlenia vysoko energetických kozmických lúčov zostávajú pre nás doposiaľ neznáme. Neutrína, ktoré k nám prilietajú prakticky bez zmeny svojho smeru a pôvodnej energie nám dokážu poodhaliť tieto zdroje v našom vesmíre. Neutrína sú ideálni nositelia informácií, ponúkajú nám unikátnu možnosť pozorovaní, ktoré ostávajú ukryté pre optické teleskopy. Princíp ich detekcie je založený na registrácii Čerenkovského žiarenia, ktoré produkujú sekundárne nabité častice vznikajúce v zrážkach neutrín s hmotou Zeme. V hlbinách vôd jazier, morí a aj v ľade sú budované tzv. neutrínové teleskopy - trojrozmerné siete fotosenzorov, ktoré sú schopné zachytiť záblesky Čerenkovského žiarenia majúceho pôvod v interakciách kozmických neutrín. Jedným z takýchto teleskopov vo výstavbe je európsky neutrínový teleskop KM3NeT (Kilometre Cubic Neutrino Telescope). Príspevok bude venovaný prezentácii teleskopu KM3NeT ako aj neutrínového tímu prof. Fedora Šimkovica z KJFB, ktorý je členom kolaborácie KM3NeT.
Non-thermal plasma with photocatalysis for decontamination of bacteria and VOC from indoor air
Thomas Vazquez, Aleksandra Lavrikova, Dalimír Wiederman, Jan Babic, and Zdenko Machala
Abstract: Long-term exposure to indoor air pollution is responsible for respiratory, cardiovascular, and oncological diseases, which is a major public health issue. Hospitals-acquired infections are also spread through air contaminants, e.g. pathogenic aerosols. Finding an innovative technology that would efficiently remove all kinds of airborne pollutants without producing harmful by-products and with a low energy cost would be not only a major advance for public health but would also help preventing the spread of airborne pathogens such as in the case of the recent COVID-19 pandemic. The goal is this work is to assess the efficacy of non-thermal plasma (NTP) combined with photocatalysis for the removal of VOCs and inactivation of aerosol-borne bacteria at a high gas flow rate. NTP and photocatalysis have proven their capabilities to decompose or inactivate a broad range of harmful compounds present in indoor air. Moreover, combining these two techniques may offer a very effective hybrid air decontamination device, as studies suggest a synergetic effect. We designed an indoor air decontamination device that combines a Dielectric Barrier Discharge (DBD) for the NTP generation, and a TiO2 coating which is activated by UV-A LEDs. Despite a very short residence time of the pollutant in the reactor (gas flow rate was set above 300 L/min), we obtained the removal efficiency of about 40%for VOC decomposition with a single-pass in the reactor. We also studied the inactivation of E.coli and S. aureus using a single-pass method. The bacteria were first prepared in solution and then sent in the reactor in a form of bioaerosols. The decontamination efficiency of these bio-aerosols are also very promising, for both bacteria, almost 100% of inactivation was reached. We also varied the duty cycle of the discharge and found very good results at a low duty cycle (15%), thus reducing the by-product ozone concentration.
This work was supported by the EU NextGenerationEU through the Recovery and Resilience Plan for Slovakia under the project No. 09I03-03-V03-00033 and Slovak Research and Development Agency APVV-22-0247.
Non-Uniform Automata: Separating Determinism, Non-determinism, and Randomization.
Pavol Duriš, Rastislav Královič, Dana Pardubská
Poster
Abstract: The non-uniform version of Turing machines with an extra advice input tape that depends on the length of the input but not the input itself is a well-studied model in complexity theory. We investigate the same notion of non-uniformity in weaker models, namely one-way finite automata. We consider the model of finite automata with advice introduced by Küçük et al. and show various relations between determinism, non-determinism, and randomization.
Numerical analysis in compressible fluid dynamics
Eduard Feireisl (Prague), Mária Lukáčová (Mainz), Hana Mizerová, Bangwei She (Beijing)
Poster
Abstract: Computational fluid dynamics provides a powerful tool to simulate the dynamics of fluids and their thermodynamical properties in a wide range of engineering applications and natural phenomena. Validation of such algorithms is typically performed experimentally and thus the question of rigorous convergence analysis for approximations of fluid motions in various settings remains open. We have proposed a new strategy to prove convergence of numerical solutions of fluid flow models based on measure-valued solutions, which has answered many previously open questions in numerical analysis of the compressible Euler and the Navier-Stokes systems with homogeneous and/or space-periodic boundary conditions. Based on the latter approach of generalized solutions we were also able to show that any uniformly bounded sequence of finite volume approximations converges unconditionally to the strong solution of the Navier-Stokes-Fourier system in the case of space-periodic boundary conditions.
Observation and characterization of titanium-like nanofilament in TiO2 memristor using superconducting electrode(s) and Andreev spectroscopy
Martin Moško, Mária Koscelanská, Antónia Mošková, Marek Vidiš, Serhii Volkov, Maroš Gregor, Magdaléna Poláčková, Tomáš Roch, Branislav Grančič, Leonid Satrapinskyy, Peter Kúš, Andrej Plecenik, Tomáš Plecenik
Abstract: A thin TiO2 semiconductor film embedded between two metal electrodes works as a memristor after being formed by soft breakdown. The forming creates a nano-filament that penetrates through the poorly conducting TiO2 film and connects the electrodes conductively. While previous works characterized the nano-filament properties (shape, composition, and resistivity) by electron microscopy techniques, we present a characterization by electrical measurements. In a typical memristor, both electrodes are made of normal metals. We study the metal/TiO2/metal memristors with a bottom electrode made of a superconducting NbN layer and a top electrode made of a normal (Pt) or superconducting (Nb) metal. The nano-filament connecting the electrodes touches the bottom electrode as a point contact, thus allowing us to perform point-contact Andreev reflection spectroscopy of the NbN superconductor. The spectra, measured below the critical temperature (15 K) of NbN, are analyzed theoretically. The analysis reveals the presence of one nano-filament and determines the nano-filament resistance, Sharvin resistance of the point contact, and Maxwell resistance of the electrodes. Moreover, it shows that the nano-filament is a conical-shaped Ti-like metal point contact with a tip diameter of ∼3–5 nm, Fermi velocity of 2x10^6 m/s, and low-temperature resistivity of ~ 10^-8 – 10^-7 Ωm. Thus, the nano-filament in our device is not the Ti4O7 phase observed in previous works. Remarkably, the point contact spectrum of the superconducting NbN layer shows the Andreev peak typical for ballistic transport. This is because the point contact probes the NbN layer through a thin Al layer that mimics superconductivity of NbN via the proximity effect and eliminates the effects of tunneling and disorder.
Odhaľovanie zdrojov svetelného znečistenia z charakteristík jasu nočnej oblohy
František Kundracik, Miroslav Kocifaj, Shweeta Soni
Abstract: Ak chceme obmedziť svetelné znečistenie v okolí miest, napríklad pri plánovaní rekonštrukcie verejného osvetlenia, je nevyhnutné identifikovať hlavné zdroje svetelného znečistenia (vysokotlakové sodíkové výbojky, LED svietidlá, osvetlenie monumentov a podobne) a urobiť predpoveď, ako sa plánovaná zmena prejaví na jase nočnej oblohy. V príspevku prezentujeme vyvinuté unikátne matematické metódy (modelovanie, analýza nameraných dát) a prístroje (automatizovaný polarimeter jasu očnej oblohy, vysokocitlivý spektrometer) vyvinuté za týmto účelom. Uvádzame tiež aktivity plánované na najbližšie obdobie.
Optická spektroskopia malých molekúl
Filip Pastierovič, Peter Čermák
Abstract: V našom laboratóriu sa zaoberáme štúdiom interakcie svetla s malými molekulami (napr. amoniak či NOx radikály). Tieto častice často zohrávajú kľúčovú úlohu v každodenných fyzikálnych a chemických procesoch v pozemskom aj vesmírnom meradle. Príkladom je charakterizácia exoplanét na základe detekcie takýchto molekúl v ich atmosférach. Na efektívnu analýzu absorpčných signálov je potrebná veľmi dobrá znalosť štruktúry týchto molekúl a ich vzájomných interakcií. Najlepšie sa tieto dáta získavajú z citlivých laboratórnych meraní, kde možno tieto referenčné dáta získať pri známych podmienkach experimentov (na rozdiel študovaných objektov, kde tieto údaje musíme zistiť z meraní). Na pracovisku sa zaoberáme ako vývojom nových optických detektorov (napr. CEAS a CRDS), tak aj spracovaním spektroskopických dát a tvorbou referenčných údajov. Z tejto oblasti je aj náš najvýznamnejší úspech – participácia na tvorbe novej verzie databázy HITRAN, ktorá je najznámejšou molekulárnou referenčnou spektroskopickou databázou.
Qualitative properties of evolution problems from science technology
prof. Michal Fečkan, prof. Pavol Quittner, a ďalší
Abstract: Poster bude zaujímavou formou prezentovať prácu jednotlivých riešiteľov tohto grantu.
Pangenome graphs improve bacterial plasmid binning in short-read assemblies
Mattia Sgro, Brona Brejova, Tomas Vinar, Paola Bonizzoni, Cedric Chauve
Abstract: Plasmid binning is a task of identifying plasmids in sequenced bacterial isolates and it is a crucial task in microbial genomics helping to monitor spread of antimicrobial resistance. Short-read genome assemblies typically consist of many contigs of variable lengths, which makes it difficult to identify sets of contigs belonging to individual plasmids. De novo methods for this problem exploit contig features such as length, coverage, circularity, or GC-content, as well as their connections in the assembly graphs. On the other hand, referenced-based strategies make use of homology to databases of known plasmids. Plasbin-flow is a hybrid method that defines plasmid bins as subgraphs of the assembly graph, identified through a MILP model that combines both contig features and plasmid database. Plasbin-flow is sensitive to the quality of the underlying genome assembly graph. In this work, we propose the use of a pangenome graph, built from assembly graphs produced by different software tools from the same sample. This pangenome graph leverages similarities between contigs from different assemblies while also retaining the information on contigs that appear only in one of the input assemblies. Our new tool first builds the pangenome graph using the ""nf-core/pangenome” pipeline, modified to retain information from input assembly graphs. It then uses a modified MILP model from Plasbin-flow to identify plasmid bins. Preliminary results on assemblies built by Unicycler and Skesa show increased recall compared to the results based on single assemblies, leading to more true plasmid contigs correctly detected.
Personalized Learning - The Student Perspective
Karolína Miková, Zuzana Kubincová
Abstract: One of the current trends in designing virtual learning environments is the emphasis on personalization and adaptability, which aims to tailor content to the needs of each individual learner. At our college, we are enhancing our educational portal to offer personalized content based on students' learning goals and current knowledge. This requires an understanding of different user requirements, which prompted us to survey our faculty and students – potential users of the portal – about their preferences for personalized learning. In this paper, we discuss the results of the student survey, highlighting the preferred learning materials, most used technologies, and desired features in a personalized virtual learning environment.
Physics Education from the view of the Theory of Five Pillars of the Mind
Daša Červeňová, Peter Demkanin
Poster
Abstract: Neurosciences significantly influence shifts in knowledge in many different fields nowadays. Results of neuroscientific studies provide explanations for many phenomena, especially the ones occurring during the learning process. Many theories proposing to promote learning, have been published. The theory of Five Pillars of the Mind, formulated by Tokuhama-Espinosa, is one of them. This article proposes an example of applying this theory to physics education. We present possible implementation of this theory and its limitations based on the first results obtained in Diploma Thesis.
PIXE analýza farieb na maliarskych plátnach
Jakub Zeman
Poster
Abstract: Particle Induced X-ray emission (PIXE) sa využíva na nedeštruktívnu analýzu rôznych vzoriek. V laboratóriu CENTA na FMFI UK boli zväzkom protónov urýchlených na vysokú energiu (3 MeV) ožarované vzorky maliarskych plátien s farbou. Toto ožarovanie je spojené s emisiou roentgenovského žiarenia. Analýzou PIXE spektier a kalibráciou spektrometra boli určené hodnoty koncentrácie jednotlivých prvkov prítomných vo farbách na maliarskych plátnach.
Possibility of Cold Atmospheric Pressure Plasma Application in Agriculture
Anna Zahoranová, Veronika Medvecká, Dušan Kováčik, Petra Šrámková, Michal Stano, Matej Klas, Sandra Ďurčányová, Martin Kuťka
Abstract: The treatment of seeds with cold atmospheric plasma (CAP) for use in agriculture has recently become a widely studied subject of multidisciplinary research. The advantage of CAP is its low-temperature nature, which allows the processing of heat-sensitive biological material, minimizes the use of harmful chemicals, and does not require vacuum equipment, but uses ambient air as a working gas. The effects of plasma treatment on plant seeds lead to an increase in germination, an acceleration of growth dynamics and plant vitality, as well as a reduction, or elimination of undesirable pathogenic microorganisms on the seed surface. The aim of our research is to study effect of CAP generated at atmospheric pressure in ambient air by several types of plasma sources relevant for application in agriculture on different types of plant seeds. We focus our research on investigating the overall effects of plasma treatment on seeds and dried fruits. Based on rich long-term cooperation with colleagues from the Department of Plant Physiology and the Department of Genetics of the Faculty of Natural Science, the impact of plasma on changes in the physicochemical properties of the seed surface, on the growth, the physiological, biochemical, molecular-biological, and anatomical aspects of seedlings after plasma treatment of seeds, as well as the potential genotoxic effects are investigated. In addition, the influence of plasma on plant resistance and the initiation of an adaptive response to environmental stressors (heavy metals, salinity, etc.), which is currently a highly current issue in the adaptation of agriculture to climate changes and environmental pollution, is being investigated. One of the goals of the presented research is to verify the effectiveness of the proposed prototype plasma device based on Diffuse Coplanar Surface Dielectric Barrier Discharge in an industrially relevant environment to improve seeds germination parameters, for which chemical treatment is standard. This method represents a promising economically and ecologically advantageous technology for the treatment of plant seeds intended for sowing, germination, storage, and consumption. This work was supported by the Slovak Research and Development Agency under the Contract no. APVV-21-0147, APVV-16-0216, and by the Slovak Grant Agency for Science VEGA 1/0688/22.
Probabilistic models for estimating virus variant proportions in wastewater samples
Askar Gafurov, Andrej Baláž, Fabian Amman, Kristína Boršová, Viktória Čabanová, Boris Klempa, Andreas Bergthaler, Tomáš Vinař, Broňa Brejová
Abstract: We developed a software tool called VirPool for analyzing data from genomic sequencing of RNA viruses extracted from wastewater. Our tool estimates proportions of known SARS-CoV-2 variants such as alpha, delta or omicron in the sample. It uses probabilistic models that capture allele frequencies in individual viral variants as well as sequencing errors and can be applied to reads from different DNA sequencing technologies. However, wastewater SARS-CoV-2 sequencing data feature high variability in the coverage of individual parts of the genome which may negatively influence such analyses. Currently, we work on modeling such artifacts in highly diluted samples.
Recent advances in tempered Psi-fractional calculus
Michal Pospíšil
Abstract: Properties of the tempered Psi-Hilfer fractional integral and the tempered Psi-Caputo fractional derivative are summarized. Recently proved Henry-Gronwall type integral inequalities and existence results for initial value problems involving the tempered Psi-Caputo fractional derivatives are presented.
Robust Self-calibration of Focal Lengths from the Fundamental Matrix
Viktor Kocur, Daniel Kyselica, Zuzana Kukelova
Abstract: The problem of self-calibration of two cameras from a given fundamental matrix is one of the basic problems in geometric computer vision. Under the assumption of known principal points and square pixels the Bougnoux formula offers a means to compute the two unknown focal lengths. However in many practical situations the formula yields inaccurate results due to commonly occurring singularities. Moreover the estimates are sensitive to noise in the computed fundamental matrix and to the assumed positions of the principal points. In this paper we therefore propose an efficient and robust iterative method to estimate the focal lengths along with the principal points of the cameras given a fundamental matrix and priors for the estimated camera intrinsics. In addition we study a computationally efficient check of models generated within RANSAC that improves the accuracy of the estimated models while reducing the total computational time. Extensive experiments on real and synthetic data show that our iterative method brings significant improvements in terms of the accuracy of the estimated focal lengths over the Bougnoux formula and other state-of-the-art methods even when relying on inaccurate priors
Rozvoj a implementácia kritického myslenia vo výučbe fyziky: Výskumné perspektívy a pedagogické inovácie
Klára Velmovská
Abstract: Rozvoj kritického myslenia je dôležitým aspektom teórie vzdelávania. Budeme prezentovať výskumy, ktoré sme v tejto oblasti realizovali. Predstavíme publikácie a aktivity určené učiteľom fyziky s cieľom implementácie rozvoja kritického myslenia do vyučovania fyziky.
Selected Problems in Applied Geometry
Marcel Makovník, Adriana Malovec Bosáková, Pavel Chalmovianský, Martina Bátorová, Barbora Pokorná
Abstract: In this paper, we present the results and ongoing research at the Department of Geometry, DAG. Special focus is on the areas of algebraic geometry, geometric modeling, geometric data processing and geometry teaching in schools.
Separators for alkaline water electrolyzers prepared by plasma-initiated grafting of acrylic acid on microporous polymeric membranes
Michal Stano, Martin Kuťka, Ľubomír Staňo, Leonid Satrapinskyy, Pavol Ďurina, Dušan Kováčik, Anna Zahoranová
Poster
Abstract: Hydrogen production by water electrolysis (WE) is expected to become a key technology in the future decarbonized economy. While the acidic proton-exchange membrane WE requires a use of precious platinum group metal electrocatalysts, alkaline WE can be realized using more abundant materials such as Ni, Co, and Fe. Advanced alkaline WE is therefore a subject of intense research and development. Energy losses and purity of product gases, H2 and O2, strongly depend on properties of a separator placed between the electrodes. We report on heterogeneous separators composed of an inert porous polymeric structure of polypropylene or polysulfone with pores filled by a hydrogel. The separators were prepared by plasma-initiated graft polymerization of acrylic acid on porous membranes, resulting in formation of poly(acrylic acid) that was converted to potassium polyacrylate by neutralization in KOH. The covalent binding of the grafted polymer with the substrate is a prerequisite for long-term stability of the produced separators. The produced separators exhibited low ionic resistance and low hydrogen permeability under electrolysis conditions in 30 wt.% KOH and 50°C. The performance of separators of various thickness was also evaluated in terms of product of ionic resistance and hydrogen permeability that is independent on membrane thickness. In terms of this parameter, the present separators surpassed the performance of conventional porous separators filled with liquid electrolyte under the identical electrolysis conditions. Stability of present separators in hot alkaline electrolyte strongly depends on conditions of the fabrication process, especially on conditions of the plasma activation. Presently, durability of separators exceeding 5000 hours has been achieved in 30 wt.% KOH and 60°C.
Short circuits and edge colourability of cubic graphs
František Kardoš, Robert Lukotka, Edita Máčajová
Abstract: We prove that each cubic graph other than Petersen graph that has each edge in a circuit of length at most 5 is 3-edge-colourable.
Spracovanie dát z 3D kamier so štrukturovaným svetlom s použitím hybdridných metód 3D počítačového videnia
Martin Madaras, Lukáš Gajdošech
Abstract: Príspevok rozoberá použitie 3D kamier so štrukturovaným svetlom v priemyselných aplikáciach a motiváciu pre využitie hybridných metód spracovania dát z týchto 3D kamier v rôznych úlohách automatizácie. Ukazuje motiváciu pre posun paradigmy v 3D počítačovom videní smerom k hlbokému učeniu, uvádza hybridnú metodológiu a diskutuje všeobecné otvorené problémy s dátovo orientovanými prístupmi. Zaoberá sa problémom nedostatku trénovacích dát na metódy hlbokého učenia pri tomto type senzorov a ako riešenie navrhuje renderovanie syntetických trenovacích dát. Navrhnuté hybridné motodológie sú potom zložené z dátovo-orientovaných modelov, ktoré su trénované na syntetických aj reálnych dátach, ako aj analytických metódach, ktoré sa použivajú ako záverečný krok spracovania. Takto hybridné metódy vedia zabezpečiť vyššiu robustnosť pri nasadení v aplikáciach a zároveň dosahovať aj vyššiu presnosť v porovnaní s inými metódami.
Textbook analysis in the MaTeK project
Zbyněk Kubáček, Peter Vankúš, Tünde Kozánek Kiss
Abstract: One of the activities of the Horizon 2020 MaTeK project was to find out what possibilities for argumentation and reasoning are provided to mathematics teachers by the textbooks used in individual participating countries. We describe the development of an appropriate theoretical framework for such an assessment.
The study of nuclear structure in collaboration with ISOLDE
J. Mišt, B. Andel, S. Antalic, A. Sitarčík
Abstract: The ISOLDE facility is the world's largest producer of radioactive ion beams, providing a great opportunity to study exotic isotopes far from the stability line. The study of these isotopes is crucial for a better understanding of atomic nuclei and their properties. Many experiments using various detection systems are performed at the facility every year. The local group at the Department of Nuclear Physics and Biophysics actively cooperates on the part of the experimental programme at the highly versatile system ISOLDE Decay Station (IDS) and worked on experiments employing the Windmill system in the past. The aim is to investigate the properties and structure of exotic atomic nuclei via combined laser and decay spectroscopy, or beta-delayed fission studies.
Topology of manifolds
Tibor Macko, Martin Niepel, Ľudovt Balko, Tomáš Rusin, Ajay Raj, Serhii Dylda, Samuel Kalužný, Marián Poturnay
Poster
Abstract: We present selected results from the work in the topology group at KAG FMFI UK supported by the grants VEGA 1/0101/17, VEGA 1/0596/21, and APVV-16-0053. The results are located in algebraic topology and deal with smooth and topological manifolds. One branch which we work in is called surgery theory. It deals with a specific way of approaching classification problems of such manifolds. We present some of the highlights of this theory, which are included in our recently published monograph. Further we present some of our own calculational results in this area dealing with specific manifolds such as lens spaces and theoretical results about the so-called rho-invariants as well as product formulas in algebraic surgery. Next we list the PhD projects of the current PhD students in our group. Another branch studied in our group deals with cohomological invariants such as the cup-length of manifolds, which is an important measure of complexity of manifolds. We present selected recent results for some classes of (oriented) Grassmann manifolds as well as flag manifolds.
Towards Explainable Malware Detection
Martin Homola, Štefan Balogh, Peter Švec, Ján Kľuka, Peter Anthony, Umberto Straccia, Franca Debole, Franco Alberto Cardillo, Ján Mojžiš, Martin Kenyeres, Monday Onoja, Zekeri Adams, Alexander Šimko
Abstract: Malware analysis has become more and more challenging due to the rapid evolution of attack techniques, and an ever increasing volume of new malware samples. Traditional approaches struggle with this load and machine learning has steadily gained ground in this application area due to its effectiveness and classification power. However, much as in other domains, malware experts lack the confidence and trust in opaque machine learning models that provide classification of the sample but little or no justifications to support and explain such decisions. Together with collaborators from APVV projects ORBIS and DyMAX and from the H2020 project TAILOR, our group is developing a suitable ontology-based data representation and a suitable set of tools based on explainable AI techniques to enable justifiable and interpretable malware detection.
Towards Excellent Robotics and Artificial Intelligence at a Slovak University
Igor Farkaš, Martin Homola, Daniela Olejárová, Viktor Kocur, Nina Zmajkovič Kučeráková
Abstract: TERAIS (Towards Excellent Robotics and Artificial Intelligence at a Slovak University) is an international effort aimed at establishing the Department of Applied Informatics (DAI) of Comenius University Bratislava, Slovakia (UKBA) as a workplace of international academic excellence. Funded by the Horizon Europe programme under the Twinning call, the project brings together experts from the University of Hamburg in Germany and the Italian Institute of Technology in Genoa to collaborate with scientists at Comenius University in Bratislava, Slovakia. The project has four key pillars, the development of people, international networking, research excellence, and research support. With a focus on cognitive robotics and artificial neural networks, the joint research efforts aim to contribute to the development of trustworthy and human-friendly robots. The project also focuses on enhancing the human capital development and research support structures at the Department of Applied Informatics (Comenius University in Bratislava, Slovakia) and building long-term international research partnerships.
Towards Integration of Photonic and Superconducting Platforms
Mykhaylo Bilogolovskyy, Magdaléna Poláčková, Olena Zhytlukhina, Maroš Gregor, Tomáš Plecenik, Oleh Feia, Andrii Shapovalov, Aaron Parra, Mariam Al Sardar, Ali Aliev
Poster
Abstract: Photonic and superconducting platforms are at the core of modern quantum technologies. Superconductors, due to their ability to maintain coherent quantum states with minimal energy loss, are essential for creating quantum computing components. Photons are ideal carriers of quantum information, offering high-speed and low-loss transmission capabilities. The integration of the two technologies poses substantial challenges because of the inherent damping of photonic states when interfacing with conventional superconductors. Transparent superconductors would be an ideal solution, minimizing photonic damping and enhancing overall system performance. Using doped n-type indium tin oxide (ITO) layers with the superconducting critical temperature up to 5 K as an example, we present design principles for such materials and our related experimental results. Magneto-transport measurements made it possible to determine temperature and angular dependences of the upper critical fields, which in a certain way demonstrate the three-dimensional nature of superconductivity in the studied samples. Point-contact conductance spectra revealed the presence, in addition to the main phase, of a phase with the critical temperature about 10 K, localized, apparently, near the surface of the ITO films. Finally, the next steps in ITO research and the future prospects for transparent superconducting electronics are discussed.
Uhlík z fosílnych a nefosílnych zdrojov v aerosóloch v Bratislave
I. Kontuľ, A. Cherkinsky, B. Bobáľ, M. Ješkovský, J. Kaizer, P. P. Povinec, M. Richtáriková, K. Sučák, J. Zeman
Abstract: Aerosóly vo všeobecnosti a uhlíkové aerosóly špecificky sú veľmi dôležité z hľadiska ich významného negatívneho vplyvu na ľudské zdravie a klimatické systémy na regionálnej aj globálnej úrovni. Preto je dôležité skúmať vlastnosti a najmä zdroje týchto častíc v atmosfére. Rádiouhlíková analýza je jednou z metód, ktoré sa v tomto výskume využívajú, a v rámci tejto štúdie sme ju aplikovali na vzorky aerosólov pravidelne odoberaných v areáli FMFI UK v Bratislave počas rokov 2022-2023. Na základe merania obsahu rádiouhlíka v týchto vzorkách pomocou urýchľovačovej hmotnostnej spektrometrie bolo možné určiť pomer častíc z fosílnych (spaľovanie fosílnych palív) a nefosílnych (spaľovanie biomasy a biogénne emisie) zdrojov. Výsledky ukázali, že väčšina uhlíka v aerosóloch v atmosfére Bratislavy pochádza z nefosílnych zdrojov. Napriek vysokej úrovni industrializácie a urbanizácie Bratislavy v priemere až 72 % uhlíka v aerosóloch je nefosílneho pôvodu, a to počas celého roka. Ich podiel sa počas skúmaného obdobia pohyboval v rozmedzí od 67 % do 82 % bez jednoznačných sezónnych variácií.
Urýchľovačová hmotnostná spektrometria na FMFI UK
Jakub Kaizer, Ivan Kontuľ, Jakub Zeman, Pavel P. Povinec
Poster
Abstract: V roku 2013 bolo na Fakulte Matematiky, Fyziky a Informatiky UK vybudované laboratórium CENTA (Centrum pre nukleárne a urýchľovačové technológie), ktoré disponuje dvomi iónovými zdrojmi, nízko energetickou separáciou iónov a tandemovým urýchľovačom Pelletron. Tento systém bol v roku 2022 doplnený o analytickú časť vysoko energetických iónov, ktorá zahŕňa analytický magnet (polomer zakrivenia 1,27 m a maximálna intenzita magnetického poľa 1,5 T), dva 45° elektrostatické analyzátory (polomer zakrivenia 2,54 m a napätie 50 kV) a mnoho anódovú ionizačnú komoru, ktorá slúži v súčasnosti ako koncový detektor. Tento systém je využívaný na urýchľovačovú hmotnostnú spektrometriu (AMS z angl. Accelerator Mass Spectrometry), ktorá patrí k najcitlivejším metódam na stanovenie dlho žijúcich rádioizotopov. V práci budú prezentované dosiahnuté výsledky z meraní 14C, 26Al a 129I ako aj separácia 10Be.
Vyhľadávanie podzemných dutín pomocou seizmického šumu
Miriam Kristeková, Jozef Kristek, Peter Moczo, Martin Gális
Poster
Abstract: Zisťovanie a lokalizácia podzemných dutín má veľký význam pri rozvoji miest a výstavbe, pretože nezistené dutiny prírodného alebo umelého pôvodu predstavujú vážne geotechnické nebezpečenstvo.
Špeciálny význam má identifikácia podzemných dutín, ktoré vznikli pri podzemných jadrových výbuchoch. Dôvodov je niekoľko. Jeden z nich je životne dôležitý a súvisí so Zmluvou o všeobecnom zákaze jadrových testov (CTBT). Zistenie a lokalizácia dutiny vytvorenej podzemným jadrovým výbuchom v rámci CTBT Inšpekcií na mieste (OSI) môže byť dôležitým dôkazom porušenia zákazu jadrových testov.
Prezentujeme dva prípady aplikácie metódy, ktorú sme nedávno vyvinuli na detekciu a lokalizáciu podzemnej dutiny - jeden pre testovacie miesto CTBT pri Felsőpetény v Maďarsku, druhý pre miesto dávneho podzemného jadrového výbuchu Tiny Tot v Nevade v USA.
Ukazujeme, že mapovanie energie zaznamenaného seizmického šumu v spektrálnej oblasti v rámci konečného intervalu frekvencií umožňuje identifikovať miesto na voľnom povrchu nad podzemnou dutinou. Metóda môže byť aplikovaná s minimálne 2 aparatúrami pomocou meraní na jednej stanici postupne presúvanej v súbore potenciálne nepravidelne rozmiestnených bodov v oblasti na voľnom povrchu Zeme nad predpokladanou dutinou a na jednej fixnej referenčnej stanici alebo pomocou súčasných meraní na celej sieti meracích bodov."
Vývoj urýchľovačových technológií pre štúdium radiačného poškodenia bio-objektov
Jakub Kaizer, Róbert Breier, Radana Gurecká, Miroslav Ješkovský, Ivan Kontuľ, Daniel Kosnáč, Ján Pánik, Pavel P. Povinec, Marta Richtáriková, Jaroslav Staníček, Karol Sučák, Michal Trnka, Jakub Zeman
Abstract: Väčšina experimentov zameraných na IBA metódy (Ion Beam Analysis) v laboratóriu CENTA (Centrum pre jadrové a urýchľovačové technológie) sa v súčasnosti vykonáva vo vákuovej komore, čo čiastočne obmedzuje analýzu špecifických vzoriek. Vzorky sú limitované veľkosťou vstupu do komory a zároveň sme limitovaní materiálom vzorky, kedy nie je možné ožarovať a analyzovať prchavé vzorky alebo vzorky, ktoré sú v tekutom stave. Toto obmedzenie je možné odstrániť systémom na extrakciu iónového zväzku do atmosféry, ktorý zvyčajne využíva tenké okienko, ktoré oddeľuje vákuum od atmosféry, pričom iónový zväzok ním prechádza so zanedbateľnými stratami. S takýmto systémom je potom možné analyzovať aj vzorky o veľkých rozmeroch (napr. obrazy a iné objekty historického dedičstva) bez poškodenia alebo ožarovať živé bunky v soľnom roztoku a sledovať ich reakciu na interakciu s iónovým zväzkom. Depozícia energie na molekule DNA a vo vrstve naviazanej vody má za následok priamo indukované primárne poškodenie DNA, ktoré je stredobodom záujmu v súvislost i s nízkymi radiačnými dávkami (LET – linear energy transfer). V poslednej dobe sa do centra záujmu dostali biologické účinky protónov ako dominantných ionizujúcich častíc v galaktickom kozmickom žiarení aj v udalostiach slnečných častíc. Preto sú jedným z hlavných zdravotných rizík pre astronautov počas dlhodobých vesmírnych misií.
WILMA as a way of connection between maths educators in universities and teachers in practice
Monika Dillingerová, Emília Miťková
Abstract: While teacher training educators at university have students in teacher training, there is a close connection between them. The students after leaving to practice, usually do not know about new research and good teaching materials made on the university. This we establish as a need of connections between us (teacher training educators) and mathematics teachers in practice. WILMA (Website of Inspirations and professional Literature for teachers of Mathematics) is a website developed by university researchers with the goal to establish this connection.
Zážitkom k poznaniu
Viera Haverlíková, Ladislav Janiga
Poster
Abstract: Neformálne vzdelávanie je bohatým zdrojom inovácií školského vzdelávania – motivácie, metód, foriem, obsahu, aj hodnotenia. Je tiež prostredím pre výskum žiackych predstáv a učebných postupov. Školské fyzikálne vzdelávanie z neho môže profitovať. V prezentácii ukážeme, ako na využívanie neformálneho vzdelávania pripravujeme budúcich učiteľov i učiteľov z praxe. Ponúkneme Vám aj konkrétnu ukážku - príďte na vlastnej koži zažiť únikovú hru Truhlica lorda Maxwella.