List your company

Adversarial Reinforcement Learning on Connect Four 100% optimal adversary (test-time)

-0.98Avg ReturnESPER
-1.0008-0.9954-0.99-0.9846Jul 25, 2024

Evaluation results

MethodLinks
ESPERtest-time adversary optimality=100%2024.07-0.98
DTtest-time adversary optimality=100%2024.07-1
ARDTtest-time adversary optimality=100%2024.07-1