AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Adaptive Online Replanning with Diffusion Models

Siyuan Zhou, Yilun Du, Shun Zhang, Mengdi Xu, Yikang Shen, Wei Xiao, Dit-Yan Yeung, Chuang Gan

NeurIPS 2023 · Reinforcement learning › Visual and representation RL · approach: RL for combinatorial routing · 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
replan with RDM instead of fixed-plan DiffuserDiffuser · Maze2D Largenormalized return: 123.0 -> 185.3Improved
replan with RDM instead of DiffuserDiffuser · Multi2D Largenormalized return: 132.1 -> 195.4Improved
replace fixed-interval replanning with likelihood-based RDMDecision Diffuser · Stochastic Hopper Med-Expertnormalized return: 49.4 -> 59.4Improved
replace fixed-interval replanning with RDMDecision Diffuser · Stochastic Walker2d Med-Expertnormalized return: 71.1 -> 92.5Improved
compare RDM with IQL on stochastic Walker2d MediumIQL · Stochastic Walker2d Mediumnormalized return: 71.8 -> 70.7Mixed
use RDM instead of Decision Diffuser on RLBenchDecision Diffuser · RLBench open boxsuccess rate: 52.9 -> 72.0Improved
use RDM instead of Decision Diffuser on RLBenchDecision Diffuser · RLBench close boxsuccess rate: 46.0 -> 61.3Improved
use RDM instead of Decision Diffuser on RLBenchDecision Diffuser · RLBench close fridgesuccess rate: 84.6 -> 98.6Improved
replan with only N=40 diffusion steps instead of Diffuser (ablation)Diffuser · Multi2D Largenormalized return: 129.4 -> 160.9Improved
increase replanning diffusion steps from N=40 to N=80 (ablation)RDM · Multi2D Largenormalized return: 160.9 -> 195.4Improved
increase replanning diffusion steps from N=80 to N=256 (replan from scratch) (ablation)RDM · Multi2D Largenormalized return: 195.4 -> 197.7No clear effect
replan on past context (RPDM) or state distance (SDM) instead of fixed intervals (DD) (ablation)Decision Diffuser · Hopper medium-expertnormalized returnImproved

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