Continuous Control on Ant v5
6,501.4Average ReturnDBC
Evaluation Results
| Method | Links | |
|---|---|---|
| DBCActor Backbone=SAC2026.02 | 6,501.4 | |
| TQCActor Backbone=SAC2026.02 | 6,342.6 | |
| SACActor Backbone=SAC2026.02 | 6,121.8 | |
| VDActor Backbone=SAC2026.02 | 3,850.4 | |
| IQNActor Backbone=SAC2026.02 | 3,487.4 | |
| VFActor Backbone=SAC2026.02 | 2,650.3 | |
| PDAEnvironment steps=1M, Seeds=10, Tests per epoch=102026.03 | 2,625.1 | |
| DSACActor Backbone=SAC2026.02 | 2,013.2 | |
| PPOEnvironment steps=1M, Seeds=10, Tests per epoch=102026.03 | 1,115.4 |