AI Paper Atlas

Research map / Reinforcement learning / Offline RL

StARformer: Transformer with State-Action-Reward Representations for Visual Reinforcement Learning

Jinghuan Shang, Kumara Kahatapitiya, Xiang Li, Michael S. Ryoo

ECCV 2022 · Reinforcement learning › Offline RL · approach: Decision Transformer sequence models · 25 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
use StARformer instead of Decision TransformerDT · Atari Breakout (offline RL)episode return: 76.9 -> 124.1Improved
use StARformer instead of Decision TransformerDT · DMC Cheetah (offline RL)episode return: 27.9 -> 275.3Improved
use StARformer instead of Decision TransformerDT · Atari Seaquest (offline RL)episode return: 1131 -> 781Hurt
use StARformer instead of Decision TransformerDT · Atari Boxing (offline RL)episode return: 78.3 -> 81.2No clear effect
use longer input sequences TStARformer · Atari and DMC (offline RL)episode returnImproved
use patch embeddings rather than conv in Step Transformer (ablation)StARformer · Atari Breakoutepisode return: 55.6 -> 124.1Improved
use conv rather than patch embeddings in Sequence Transformer (ablation)StARformer · Atari Breakoutepisode return: 38.0 -> 124.1Improved
add state embedding (conv) in Sequence Transformer (ablation)StARformer · Atari Breakoutepisode return: 38.7 -> 124.1Improved
use layer-wise attention connection instead of summation fusion (ablation)StARformer · Atari Breakoutepisode return: 29.2 -> 124.1Improved
use layer-wise connection instead of stacking (ablation)StARformer · Atari Qbertepisode return: 575 -> 6968Improved
replace conv state embedding by ViT in DT (ablation)DT · Atari Breakoutepisode return: 76.9 -> 47.3Hurt
use stepwise reward instead of return-to-go (ablation)StARformer · Atari and DMCepisode returnNo clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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