List your company
ResearchFormerly WizWand

Reinforcement Learning on PointMaze Medium v3

1.06Avg AUC (z-scored)SAC
-0.04240.24380.530.8162Jun 15, 2026
Updated 4mo ago

Evaluation results

MethodLinks
SACNoise Type=Red2026.061.06
TD3Noise Type=Red2026.060.82
TD3Noise Type=OU2026.060.7
TD3Noise Type=Pink2026.060.66
SACNoise Type=Blush2026.060.49
SACNoise Type=Pink2026.060.41
TD3Noise Type=Baby2026.060.27
TD3Noise Type=Rose2026.060.26
SACNoise Type=Rose2026.060.23
TD3Noise Type=Blush2026.060.19
SACNoise Type=White2026.060.08
TD3Noise Type=White2026.060.06
SACNoise Type=Baby2026.060