AI Paper Atlas

Research map / Reinforcement learning / Offline RL

MetaDiffuser: Diffusion Model as Conditional Planner for Offline Meta-RL

Fei Ni, Jianye HAO, Yao Mu, Yifu Yuan, YAN ZHENG, Bin Wang, Zhixuan Liang

ICML 2023 · Reinforcement learning › Offline RL · approach: Diffusion models for offline decision making · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use task-conditioned diffusion planning instead of the strongest listed baselineOffline meta-RL · Ant-Diraverage test return: 213.2 -> 247.7Improved
Use task-conditioned diffusion planning instead of the strongest listed baselineOffline meta-RL · Walker-Paramaverage test return: 300.5 -> 368.3Improved
Use task-conditioned diffusion planning instead of the strongest listed baselineOffline meta-RL · Point-Robot, Cheetah-Dir, Cheetah-Vel, Hopper-Paramaverage test return: -5.04 -> -4.48; 931.7 -> 936.2; -51.3 -> -45.9; 289.3 -> 356.4Improved
Add dynamics guidance alongside reward guidance (ablation)MetaDiffuser · Ant-Dir, Cheetah-Vel, Walker-Param, Hopper-Paramaverage test return: 210.3 -> 247.7; -52.9 -> -45.9; 326.5 -> 368.3; 293.3 -> 356.8Improved
Add dynamics guidance alongside reward guidance (ablation)MetaDiffuser · Point-Robot, Cheetah-Diraverage test return: -4.57 -> -4.48; 924.2 -> 936.2Mixed
Replace conditional diffusion with a conditioned VAE planner (ablation)MetaDiffuser · MuJoCo and Point-Robot tasksaverage test return: -4.48 -> -6.11; 247.7 -> 130.6; 936.2 -> 759.4; -45.9 -> -87.1; 368.3 -> 187.9; 356.4 -> 157.3Hurt
Remove task conditioning from the diffusion model (ablation)MetaDiffuser · MuJoCo and Point-Robot tasksaverage test return: -4.48 -> -7.48; 247.7 -> 53.2; 936.2 -> 614.7; -45.9 -> -121.7; 368.3 -> 135.2; 356.4 -> 103.5Hurt
Use CORRO task representations instead of the paper's context representation (ablation)MetaDiffuser · MuJoCo and Point-Robot tasksaverage test returnMixed
Use ground-truth task parameters as context (ablation)MetaDiffuser · MuJoCo and Point-Robot tasksaverage test returnImproved
Use medium or random warm-start data instead of expert data (ablation)MetaDiffuser · Six reported environmentsaverage test return: Expert -> Random: -4.48 -> -4.61; 247.7 -> 213.8; 936.2 -> 916.7; -45.9 -> -55.8; 368.3 -> 341.6; 356.4 -> 319.6Hurt
Use medium or random warm-start data instead of expert data (ablation)Prompt-DT · Ant-Dir, Cheetah-Vel, Walker-Param, Hopper-Paramaverage test return: 213.2 -> 40.1; -51.3 -> -208.4; 287.7 -> 64.7; 265.2 -> 82.6Hurt
Train the context encoder on mixed expert and random data instead of expert data (ablation)MetaDiffuser · Ant-Dir, Cheetah-Vel, Walker-Param, Hopper-Paramaverage test return: 247.7 -> 258.4; -45.9 -> -43.4; 368.3 -> 381.5; 356.4 -> 375.2Improved

Later papers that compared with this method

1 later accepted paper reports MetaDiffuser as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy