pen
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Off-policy evaluation for classification error | pen | Bias-0.095 | 15 | |
| Offline-to-online Reinforcement Learning | pen | Regret5.3 | 12 | |
| Trojan Attack (Target action: 'fixed random') | Pen | Attack Success Rate (ASR)100 | 9 | |
| Trojan Attack (Target action: 'arithmetic') | Pen | ASR0 | 9 | |
| Trojan Attack (Target action: '1') | Pen | ASR100 | 9 | |
| Reinforcement Learning | pen human | Normalized Return53.4 | 4 | |
| Reinforcement Learning | pen cloned | Normalized Return58.9 | 4 |
Showing 7 of 7 rows