Atari Pong
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Reinforcement Learning | Atari Pong | Mean Episode Return21 | 19 | |
| Backdoor Attack Detection | Atari Pong | AUROC87.9 | 11 | |
| Offline Reinforcement Learning | Atari Pong | Episode Return18.8 | 6 | |
| Backdoor Defense Performance | Atari Pong Clean Environment | Score1 | 5 | |
| Backdoor Defense Performance | Atari Pong Poisoned Environment | Defense Score0.973 | 5 | |
| Offline Reinforcement Learning | Atari Pong 1% DQN-replay v0 | Gamer-Normalized Score111.9 | 5 | |
| Discrete Control | Atari Pong NoFrameskip v4 | Total Reward20.6 | 3 | |
| Imitation Learning | Atari Pong v4 (test) | Reward21 | 2 |
Showing 8 of 8 rows