AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Episodic Memory-Guided Controllable Experience Synthesis for Reinforcement Learning

Xiao Ma, Tian Li, Wu-Jun Li

ICML 2026 · Reinforcement learning › Visual and representation RL · approach: Probabilistic and autonomous RL

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
condition the diffusion model on both h(s) and EM-estimated return r + gamma e(h(s')) rather than either alone (ablation)EMCES · Hopper medexp, TD3+BCnormalized score: 91.7 / 91.0 -> 105.5Improved
use EM-estimated r + gamma e(h(s')) instead of discounted return d(s) as condition (ablation)EMCES · Hopper medexp, TD3+BCnormalized score: 100.9 -> 105.5Improved
use EMTD-error prioritized condition sampling instead of uniform (ablation)EMCES · Hopper medexp, TD3+BCnormalized score: 91.6 -> 105.5Improved
use EMTD-error sampling instead of return-based or reward-based sampling (ablation)EMCES · Hopper medexp, TD3+BCnormalized score: 95.2 (return) / 99.4 (reward) / 103.6 (e(h(s))) -> 105.5Improved
use hashing-based state representation instead of grid-based (ablation)EMCES · Hopper medexp, TD3+BCnormalized score: 98.2 -> 105.5Improved
use hashing-based state representation instead of random projection (ablation)EMCES · Hopper medexp, TD3+BCnormalized score: 104.0 -> 105.5; storage 7806.49 -> 0.96 KB, time 7573.88 -> 297.12 sNo clear effect
replace original data with EMCES synthetic data of same sizeTD3+BC · D4RL gym MuJoCo sumsum of normalized scores: 683.5 (SynthER) -> 731.4Improved
use EMCES synthetic data instead of SynthER for IQLIQL · D4RL hopper-medium-replaynormalized score: 42.3 -> 101.0Improved
EMCES synthetic data vs original dataset for EDACEDAC · D4RL gym MuJoCo sumsum of normalized scores: 837.7 (original) vs 837.0No clear effect
use EMCES synthetic data on vision-based VD4RL (latent space)DrQ+BC · VD4RL walker-walk and cheetah-run averageaverage episode reward: 472.38 -> 518.00Improved
use SAC (EMCES) instead of SAC (SynthER) and SAC (PGR) in online RLSAC · online DMC and Gym (6 environments)mean average 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