Deep Reinforcement Learning on Arcade Learning Environment 100K (test)
0.0927Median Human Normalized ScoreCoAct TD
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| CoAct TDBase Algorithm=DDQN2026.03 | 0.0927 | 0.0145 | 0.3762 | |
| NoisyNetworksBase Algorithm=DDQN2026.03 | 0.0457 | 0.0102 | 0.1913 | |
| ϵ-greedyBase Algorithm=DDQN2026.03 | 0.0377 | 0.0056 | 0.2942 |