Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Single-agent Reinforcement Learning (SARL) on Breakout Poisoned
Loading...
570.9
Average Episode Return
SHINE
1.292
149.171
297.05
444.929
May 7, 2026
Average Episode Return
Updated 26d ago
Evaluation Results
Method
Method
Links
Average Episode Return
SHINE
2026.05
570.9
BehaviorGuard
2026.05
517.8
STRIP
2026.05
444.6
BIRD
2026.05
404
Direct retraining
2026.05
309
PD
2026.05
220.7
Original
2026.05
33.1
FeatureRE
2026.05
31
NC
2026.05
23.2
Feedback
Search any
task
Search any
task