Deep Reinforcement Learning on Q*bert Atari 2600
8,160IQM ReturnSpectral Routing
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Spectral RoutingReplay Strategy=Spectral Routing, Compute-Match status=false, Training Time (hr)=5.42026.04 | 8,160 | 35 | |
| REMReplay Strategy=Random Ensemble Mixture, Compute-Match status=false2026.04 | 7,870 | — | |
| PERReplay Strategy=Prioritized Experience Replay, Compute-Match status=false2026.04 | 7,750 | — | |
| Uniform ReplayReplay Strategy=Uniform, Compute-Match status=true2026.04 | 7,595 | — |