Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Continual Reinforcement Learning on CoinRun Normalized Continual Learning
Loading...
1.34
Max Performance
ARROW
0.8304
0.9627
1.095
1.2273
Mar 12, 2026
Max Performance
Environment Frames (Median)
Success Rate (Runs >= 85%)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Max Performance
Environment Frames (Median)
Success Rate (Runs >= 85%)
ARROW
Task Order=Reversed Ta...
2026.03
1.34
3,768,320
100
ARROW
Task Order=Two-Cycle T...
2026.03
1.33
3,768,320
100
DV3
Task Order=Reversed Ta...
2026.03
1.28
3,604,480
100
DV3
Task Order=Two-Cycle T...
2026.03
1.23
1,802,240
100
ARROW
Task Order=Default Tas...
2026.03
1.19
1,638,400
100
DV3
Task Order=Default Tas...
2026.03
1.19
491,520
100
TES-SAC
Task Order=Two-Cycle T...
2026.03
0.9
-
0
TES-SAC
Task Order=Default Tas...
2026.03
0.86
-
0
TES-SAC
Task Order=Reversed Ta...
2026.03
0.85
-
0
Feedback
Search any
task
Search any
task
