List your company
ResearchFormerly WizWand

Imitation Learning on Maze

94.4Success RateDP
77.34481.77286.290.628Jul 1, 2026
Updated 4mo ago

Evaluation results

MethodLinks
DPPolicy Architecture=Diffusion2026.0794.4
LC-DP (OURS)Policy Architecture=Diffusion2026.0794.4
LC-DP-REWARDPolicy Architecture=Diffusion2026.0794
LPB-OFFLINEPolicy Architecture=Diffusion2026.0793.6
LC-DP-CLASSIFIERPolicy Architecture=Diffusion2026.0792.4
DWBCPolicy Architecture=Feedforward2026.0785.2
ILIDPolicy Architecture=Feedforward2026.0782.4
LC-BC (OURS)Policy Architecture=Feedforward2026.0781.6
BCPolicy Architecture=Feedforward2026.0779.6
LC-BC-REWARDPolicy Architecture=Feedforward2026.0779.6
DEMODICEPolicy Architecture=Feedforward2026.0778
LC-BC-CLASSIFIERPolicy Architecture=Feedforward2026.0778