List your company
ResearchFormerly WizWand

Continuous Control on MuJoCo HalfCheetah v4

107Normalized PerformanceAD-SAC
-3.2425.385482.62May 23, 2024Sep 29, 2024Feb 5, 2025Jun 14, 2025Oct 21, 2025Feb 27, 2026Jul 6, 2026
Updated 3mo ago

Evaluation results

MethodLinks
AD-SACDelays=5, Global steps=1M2024.05107
VDPODelays=5, Global steps=1M2024.05103
BPQLDelays=5, Global steps=1M2024.05100
DIDADelays=5, Global steps=1M2024.0590
BPQLDelays=25, Global steps=1M2024.0587
AD-SACDelays=50, Global steps=1M2024.0574
BPQLDelays=50, Global steps=1M2024.0573
VDPODelays=50, Global steps=1M2024.0572
AD-SACDelays=25, Global steps=1M2024.0571
VDPODelays=25, Global steps=1M2024.0570
DC/ACDelays=5, Global steps=1M2024.0540
A-SACDelays=5, Global steps=1M2024.0535
DC/ACDelays=25, Global steps=1M2024.0516
DUPOMaximum observation delay (ΔTmax)=25, Training steps=1M2026.0715.96
DIDADelays=50, Global steps=1M2024.0515
State PredictionMaximum observation delay (ΔTmax)=25, Training steps=1M2026.0713.19
DIDADelays=25, Global steps=1M2024.0512
A-SACDelays=50, Global steps=1M2024.0512
DC/ACDelays=50, Global steps=1M2024.0512
DUPOMaximum observation delay (ΔTmax)=10, Training steps=1M2026.077.53
State PredictionMaximum observation delay (ΔTmax)=10, Training steps=1M2026.076.42
State AugmentationMaximum observation delay (ΔTmax)=10, Training steps=1M2026.075.96
State AugmentationMaximum observation delay (ΔTmax)=25, Training steps=1M2026.075.1
DUPOMaximum observation delay (ΔTmax)=5, Training steps=1M2026.074.53
A-SACDelays=25, Global steps=1M2024.054
State PredictionMaximum observation delay (ΔTmax)=5, Training steps=1M2026.073.93
State AugmentationMaximum observation delay (ΔTmax)=5, Training steps=1M2026.073.78
DC/ACMaximum observation delay (ΔTmax)=5, Training steps=1M2026.071
DC/ACMaximum observation delay (ΔTmax)=10, Training steps=1M2026.071
DC/ACMaximum observation delay (ΔTmax)=25, Training steps=1M2026.071