AI Paper Atlas

Research map / Reinforcement learning / Offline RL

Mutual Information Regularized Offline Reinforcement Learning

Xiao Ma, Bingyi Kang, Zhongwen Xu, Min Lin, Shuicheng YAN

NeurIPS 2023 · Reinforcement learning › Offline RL · approach: Behavior regularization for offline RL · 1 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
Increase Monte-Carlo samples from 5 to 50 (ablation)MISA · gym-locomotion-v2 (total)normalized score: 711 -> 742.9Improved
Increase Monte-Carlo samples from 20 to 50 (ablation)MISA · gym-locomotion-v2 (total)normalized score: 721.6 -> 742.9Improved
Increase MCMC burn-in steps from 1 to 5 (ablation)MISA · gym-locomotion-v2 (total)normalized score: 687.2 -> 742.9Improved
Add the Barber-Agakov term (ablation)MISA · gym-locomotion-v2 (total)normalized score: 715 -> 742.9Improved
Use tighter MI bound than Barber-Agakov bound alone (ablation)MISA · gym-locomotion-v2 (total)normalized score: 254.4 -> 742.9Improved
Use tighter bound than f-divergence variant (ablation)MISA · gym-locomotion-v2 (total)normalized score: 504.7 -> 742.9Improved
Use tighter bound than Donsker-Varadhan variant (ablation)MISA · gym-locomotion-v2 (total)normalized score: 673.7 -> 742.9Improved
Add bias correction with extra MCMC samples (ablation)MISA · gym-locomotion-v2 (total)normalized score: 720.9 -> 742.9Improved
Use Q function as energy instead of separate T network (ablation)MISA · gym-locomotion-v2 (total)normalized score: 271.0 -> 742.9Improved
Learn separate T network (MISA-T) instead of Q (ablation)MISA · halfcheetah-medium-v2normalized score: 60.8 vs 47.4Hurt
Remove Barber-Agakov term (ablation)MISA · hopper-medium-replay-v2normalized score: 98.6 -> 100.3No clear effect
MISA instead of IQL on antmazeIQL · antmaze-v0 (total)normalized score: 378 -> 348.1Hurt

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