List your company

Offline Reinforcement Learning on OGBench cube-triple (ct)

24Success RateIQN
-0.965.521218.48May 7, 2026May 8, 2026

Evaluation results

MethodLinks
IQNPolicy Type=Gaussian Policies2026.0524
Value FlowsPolicy Type=Flow Policies2026.0520
FlowIQN-FQLPolicy Type=Flow Policies2026.0510
CGQLMethod Category=GUIDANCE, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.058
CGQL-LMethod Category=GUIDANCE, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.058
CGQL-MMethod Category=GUIDANCE, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.058
QAM-EMethod Category=ADJOINT MATCHING, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.055
ME-AMMethod Category=ADJOINT MATCHING, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.055
FlowIQN-RPolicy Type=Flow Policies2026.055
DACMethod Category=GUIDANCE, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.054
BAMMethod Category=ADJOINT MATCHING, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.054
QAMMethod Category=ADJOINT MATCHING, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.054
FQLPolicy Type=Gaussian Policies2026.054
FQLMethod Category=BACKPROP, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.053
FEditMethod Category=BACKPROP, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.053
QSMMethod Category=GUIDANCE, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.053
QAM-FMethod Category=ADJOINT MATCHING, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.052
CODACPolicy Type=Gaussian Policies2026.052
ReBRACMethod Category=GAUSSIAN, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.051
DSRLMethod Category=POST-PROCESSING, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.051
FAWACMethod Category=GAUSSIAN, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.050
FBRACMethod Category=BACKPROP, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.050
IFQLMethod Category=POST-PROCESSING, Training Steps=1M, Seeds=8, Chunk Size (h)=52026.050
BCPolicy Type=Gaussian Policies2026.050
ReBRACPolicy Type=Gaussian Policies2026.050