Navigation on MiniWorld Sidewalk (final 200 train episodes)
44.3Episode LengthPPO + Stimuli
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| PPO + StimuliReward Regime=Baseline2026.05 | 44.3 | 21.81 | 0 | |
| PPO + VCReward Regime=Intrinsic Reward Only, theta_thr=0.952026.05 | 100.23 | 16.49 | 0 | |
| Base PPOReward Regime=Baseline2026.05 | 116.88 | 0 | 0 | |
| PPO + VCReward Regime=Composite Reward (Extrinsic + Intrinsic), theta_thr=0.252026.05 | 126.38 | 5.71 | 0 | |
| PPO + VCReward Regime=Intrinsic Reward Only, theta_thr=0.252026.05 | 130.59 | 2.09 | 0 | |
| PPO + VCReward Regime=Composite Reward (Extrinsic + Intrinsic), theta_thr=0.952026.05 | 133.8 | 1.84 | 0 | |
| PPO + VCReward Regime=Intrinsic Reward Only, theta_thr=0.602026.05 | 139.96 | 1.73 | 0 | |
| PPO + VCReward Regime=Composite Reward (Extrinsic + Intrinsic), theta_thr=0.602026.05 | 142.16 | 1.43 | 0 |