Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Simultaneous Optimization on HybridCoop+AllD Flagship setting (full-cooperation reference 2.25)
Loading...
2.24
Per-interaction Payoff
Reciprocity gradient
1.512
1.701
1.89
2.079
May 8, 2026
Per-interaction Payoff
Reference Achievement (%)
Both-Discriminator Seed Success Rate (%)
Updated 22d ago
Evaluation Results
Method
Method
Links
Per-interaction Payoff
Reference Achievement (%)
Both-Discriminator Seed Success Rate (%)
Reciprocity gradient
Opponent access=oracle...
2026.05
2.24
99
4
Reciprocity gradient
Opponent access=learne...
2026.05
2.225
99
20
DPG
Gradient path=sampled...
2026.05
1.81
80
0
DDPG
Gradient path=sampled...
2026.05
1.55
69
0
TD3
Gradient path=sampled...
2026.05
1.54
69
0
Feedback
Search any
task
Search any
task