List your company

Opponent Shaping on Rescue-the-General (RTG) (test)

20.5Win RateD-BOS
4.0688.33412.616.866May 27, 2026

Evaluation results

MethodLinks
D-BOSk=32026.0520.5
D-BOSk=12026.0518.5
PPOshaping=none2026.0510.1
D-BOSk=52026.056.6
BBMreward=per-timestep Bayes-factor intrinsic reward2026.054.7