AI Paper Atlas

Research map / Reinforcement learning / Offline RL

Train Once, Get a Family: State-Adaptive Balances for Offline-to-Online Reinforcement Learning

Shenzhi Wang, Qisen Yang, Jiawei Gao, Matthieu Gaetan Lin, HAO CHEN, Liwei Wu, Ning Jia, Shiji Song, Gao Huang

NeurIPS 2023 · Reinforcement learning › Offline RL · approach: Offline-to-online fine-tuning · 4 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
add FamO2O state-adaptive balance to IQLIQL · D4RL Locomotion totalnormalized score: 718.3 -> 772.0Improved
add FamO2O to IQLIQL · D4RL AntMaze totalnormalized score: 428.7 -> 478.3Improved
add FamO2O to AWACAWAC · D4RL Locomotion totalnormalized score: 648.4 -> 714.9Improved
add FamO2O to AWACAWAC · D4RL AntMaze totalnormalized score: 124.7 -> 243.7Improved
add FamO2O to AWACAWAC · antmaze-medium-playnormalized score: 0.0 -> 34.2Improved
add FamO2O to IQLIQL · antmaze-umaze-diversenormalized score: 37.8 -> 70.8Improved
add FamO2O to AWACAWAC · hopper-medium-expertnormalized score: 97.7 -> 92.9Hurt
add FamO2O to IQLIQL · halfcheetah-medium-replaynormalized score: 53.7 -> 53.1No clear effect
add FamO2O to AWAC and IQL on medium datasets with consistent data qualityAWAC/IQL · hopper, halfcheetah, walker2d mediumnormalized score: 65.4 -> 90.7Improved
replace balance model with random balance coefficient selector (ablation)FamO2O · D4RL Locomotion totalnormalized score: 771.96 -> 730.70Hurt
replace balance model with random selector (ablation)FamO2O · hopper-medium-replaynormalized score: 97.64 -> 80.87Hurt
replace balance model with random selector (ablation)FamO2O · halfcheetah-medium-replaynormalized score: 53.07 -> 53.75No clear effect

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