Continuous Control on HalfCheetah v5
13,996.2Average ReturnTQC
Evaluation Results
| Method | Links | |
|---|---|---|
| TQCActor Backbone=SAC2026.02 | 13,996.2 | |
| DBCActor Backbone=TD32026.02 | 13,787.8 | |
| DSACActor Backbone=SAC2026.02 | 12,974 | |
| SACActor Backbone=SAC2026.02 | 12,776.4 | |
| IQNActor Backbone=SAC2026.02 | 10,930.2 | |
| VFActor Backbone=SAC2026.02 | 8,199.5 | |
| PDAEnvironment steps=1M, Seeds=10, Tests per epoch=102026.03 | 5,174.6 | |
| VDActor Backbone=SAC2026.02 | 4,284.8 | |
| PPOEnvironment steps=1M, Seeds=10, Tests per epoch=102026.03 | 4,067.5 |