AI Paper Atlas

Research map / Reinforcement learning / Offline RL

LAPO: Latent-Variable Advantage-Weighted Policy Optimization for Offline Reinforcement Learning

Xi Chen, Ali Ghadirzadeh, Tianhe Yu, Jianhao Wang, Yuan Gao, Wenzhe Li, Liang Bin, Chelsea Finn, Chongjie Zhang

NeurIPS 2022 · Reinforcement learning › Offline RL · approach: Offline RL sample complexity theory · 5 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
LAPO latent generative model vs PLAS VAEoffline RL policy · Toy navigation taskaverage return: 1.03 -> 4.24Improved
replace Gaussian with GMM policy in AWACAWAC · Toy navigation taskaverage return: 2.43 -> 4.39Improved
replace Gaussian with GMM policy in AWACAWAC · Walker2d-mix-forward-v1normalized score: 71.89 -> 78.09Improved
replace AWAC Gaussian with GMM policyAWAC · Maze2d-umaze-v1normalized score: 94.53 -> 19.45Hurt
use LAPO instead of next best baselineLAPO · heterogeneous tasks (12)normalized score: 49% average improvementImproved
use LAPO on narrow/biased locomotion datasetsLAPO · D4RL locomotion random/medium/expertnormalized scoreMixed
LAPO vs CQL on Hopper-mediumLAPO · Hopper-medium-v2normalized score: 71.59 -> 51.63Hurt
LAPO vs best baseline on Hopper-randomLAPO · Hopper-random-v2normalized score: 8.33 -> 23.46Improved
remove latent policy training (sample z from prior) (ablation)LAPO · Walker2D-medium and Walker2d-mix-backwardnormalized scoreNo clear effect
remove latent policy training (sample z from prior) (ablation)LAPO · Maze2D-medium and Antmaze-medium-diversenormalized scoreHurt
remove limit on latent z range [-zmax, zmax] (ablation)LAPO · four tasksnormalized scoreHurt
LAPO vs best baseline on Maze2d-largeLAPO · Maze2d-large-v1normalized score: 64.30 -> 200.56Improved

Later papers that compared with this method

2 later accepted papers report LAPO 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