Reinforcement Learning on Cheetah-Wind-E dynamics changes episodic
-41.6Average ReturnAda-Diffuser + IDQL (Oracle)
Evaluation Results
| Method | Links | |
|---|---|---|
| Ada-Diffuser + IDQL (Oracle)Base Algorithm=IDQL, Latent Factor Integration=Ada-Diffuser, Oracle Status=true2026.05 | -41.6 | |
| Ada-Diffuser + IDQLBase Algorithm=IDQL, Latent Factor Integration=Ada-Diffuser, Oracle Status=false2026.05 | -48.5 | |
| Ada-Diffuser + DP (Oracle)Base Algorithm=DP, Latent Factor Integration=Ada-Diffuser, Oracle Status=true2026.05 | -52 | |
| Ada-Diffuser + DPBase Algorithm=DP, Latent Factor Integration=Ada-Diffuser, Oracle Status=false2026.05 | -58.5 | |
| IDQL + DynaMITEBase Algorithm=IDQL, Latent Factor Integration=DynaMITE, Oracle Status=false2026.05 | -59 | |
| DP + DynaMITEBase Algorithm=DP, Latent Factor Integration=DynaMITE, Oracle Status=false2026.05 | -72.2 | |
| IDQLBase Algorithm=IDQL, Latent Factor Integration=None, Oracle Status=false2026.05 | -97.5 | |
| DPBase Algorithm=DP, Latent Factor Integration=None, Oracle Status=false2026.05 | -104.8 |