Algorithmic Learning Theory - Sink Equilibria, Game Convergence, and Data-Dependent Learning - Session 2

Explore five cutting-edge research presentations from the 37th International Conference on Algorithmic Learning Theory covering fundamental advances in game theory, online learning, and machine learning paradigms. Discover how sink equilibria function as attractors in learning games, examine last-iterate convergence properties for symmetric general-sum games under exponential weights dynamics, and learn about strategy-robust approaches to contextual pricing problems. Investigate a novel data-dependent learning framework designed for large hypothesis classes and understand the inherent limitations of Empirical Risk Minimization when applied to synthetic data. Each presentation delivers theoretical insights and practical implications for algorithmic learning theory, featuring contributions from leading researchers in computational learning and game theory.