Run the GAMUT: A Comprehensive Approach to Evaluating Game-Theoretic Algorithms

We present GAMUT1 , a suite of game generators designed for testing game-theoretic algorithms. We explain why such a generator is necessary, offer a way of visualizing relationships between the sets of games supported by GAMUT, and give an overview of GAMUT’s architecture. We highlight the importance of using comprehensive test data by benchmarking existing algorithms. We show surprisingly large variation in algorithm performance across different sets of games for two widely-studied problems: computing Nash equilibria and multiagent learning in repeated games.2

• Type: Conference
• Year: 2004
• Where: ATAL
• Authors: Eugene Nudelman, Jennifer Wortman, Yoav Shoham, Kevin Leyton-Brown
• Keywords: ATAL 2004, Comprehensive Test Data, GAMUT’s Architecture, Testing Game-theoretic Algorithms