Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Navigation on Safety Gymnasium Push (test)
Loading...
4.77
Return
PPO-Lag
0.922
1.921
2.92
3.919
Mar 24, 2026
Return
Cost
Learned Cost
Accuracy
Updated 24d ago
Evaluation Results
Method
Method
Links
Return
Cost
Learned Cost
Accuracy
PPO-Lag
Threshold=35, Oracle=true
2026.03
4.77
36.17
-
-
PPO-BT
Threshold=35
2026.03
4.09
41.85
1.62
84
PbCRL
Threshold=35
2026.03
3.2
34.86
0.73
86
RLSF
Threshold=35
2026.03
1.07
20.05
80.48
80
Feedback
Search any
task
Search any
task
