List your company

Nash Equilibrium Learning on Offline Two-Player Zero-Sum Markov Games

1Statistical RatePNVI
0.950.97511.025May 13, 2026

Evaluation results

MethodLinks
PNVIKL-reg.=✗, Multi-step=✓2026.051
PMVIKL-reg.=✗, Multi-step=✓2026.051
BCELKL-reg.=✗, Multi-step=✓2026.051
Ye et al.KL-reg.=✓, Multi-step=✗2026.051
SOS-MDKL-reg.=✓, Multi-step=✓2026.051