Seminar of Theoretical Computer Science - Maciej Skorski (1.3.2019)

Friday 1.3.2019 at 11:00, room M/213

26. 02. 2019 12.05 hod.
By: Rastislav Kráľovič

Maciej Skorski:
Approximating Min-Max Strategies by Statistical Learning

Abstract:

Finite approximation of mixed strategies is crucial for many applications of the von Neumann min-max theorem. I will show how this problem is related to generalization bounds from machine learning, and discuss applications to game-theory, convex approximation and cryptography.