List your company

Reinforcement Learning on Ant CrossMaze

1,010.04RewardDTSemNets
188.6584401.9017615.145828.3883May 18, 2026

Evaluation results

MethodLinks
DTSemNets2026.051,010.04
VIPER (PPO)2026.05996.54
PPO2026.05657.03
π-PRLPolicy Type=Discretized2026.05506.32
π-PRLPolicy Type=Continuous relaxed2026.05363.44
π-PRLPolicy Type=Fine-tuned2026.05223.58
DiPRL2026.05220.25