AI Paper Atlas

Research map / Reinforcement learning / Offline RL

You Can’t Count on Luck: Why Decision Transformers and RvS Fail in Stochastic Environments

Keiran Paster, Sheila A. McIlraith, Jimmy Ba

NeurIPS 2022 · Reinforcement learning › Offline RL · approach: Decision Transformer sequence models · 8 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
Replace conditioning on sampled trajectory returns with learned cluster-average returnsRvS / Decision Transformer · Gambling, Connect Four, and 2048alignment between target and achieved returnImproved
Replace conditioning on sampled trajectory returns with learned cluster-average returnsRvS / Decision Transformer · Connect Fouraverage return: 0.2 -> win and lose with perfect accuracyImproved
Replace conditioning on sampled trajectory returns with learned cluster-average returnsRvS / Decision Transformer · 2048win rate: about $60\%$ -> $30\%$ to $80\%$Improved
Replace conditioning on sampled trajectory returns with learned cluster-average returnsRvS / Decision Transformer · Gamblingmaximum expected return: cannot get positive reward -> 1.0Improved
Train on more offline dataReturn-conditioned RvS · 2048alignment between target and achieved return: 5\% -> 100\%No clear effect
Train on more offline dataESPER · 2048alignment between target and achieved return: 5\% -> 100\%Improved
Use trajectory statistics that allow the dynamics model to exploit stochastic outcomes (ablation)ESPER clustering and policy · Gamblingagent performanceHurt
Use an MLP policy instead of a transformer policy (ablation)ESPER · Stochastic benchmark tasksperformanceNo clear effect
Use ESPER rather than return-conditioned Decision Transformer (ablation)RvS · D4RL Mujoco tasksperformanceNo 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