List your company

Adversarial Reinforcement Learning on Connect Four 50% optimal adversary (test-time)

0.11Average ReturnARDT
-0.2956-0.1903-0.0850.0203Jul 25, 2024

Evaluation results

MethodLinks
ARDTtest-time adversary optimality=50%2024.070.11
ESPERtest-time adversary optimality=50%2024.070
DTtest-time adversary optimality=50%2024.07-0.28