AI Paper Atlas

Research map / Reinforcement learning / Offline RL

StARformer: Transformer With State-Action-Reward Representations for Robot Learning

Jinghuan Shang, Xiang Li, Kumara Kahatapitiya, Yu‐Cheol Lee, Michael S. Ryoo

TPAMI 2022 · Reinforcement learning › Offline RL · approach: Decision Transformer sequence models · 18 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 StAR-representations via Step TransformerDT · Atari Breakout (offline RL)episodic return: 76.9 -> 124.1Improved
add StAR-representations via Step TransformerDT · DMC Cheetah (offline RL)episodic return: 27.9 -> 275.3Improved
add StAR-representations via Step TransformerDT · Atari Seaquest (offline RL)episodic return: 1131 -> 781Hurt
add StAR-representations via Step TransformerDT · Atari Boxing (offline RL)episodic return: 78.3 -> 81.2No clear effect
add StAR-representations via Step TransformerDT · Atari Breakout (imitation)episodic return: 54.3 -> 103.1Improved
add StAR-representations via Step TransformerDT · DMC Cheetah (imitation)episodic return: 0.5 -> 231.9Improved
use patch embeddings in Step Transformer instead of conv (ablation)StARformer · Atari Breakout (offline RL)episodic return: 55.6 -> 124.1Improved
use conv features for pure state tokens instead of patches (ablation)StARformer · Atari Breakout (offline RL)episodic return: 38.0 -> 124.1Improved
add pure conv state tokens in Sequence Transformer (ablation)StARformer · Atari Breakout (offline RL)episodic return: 38.7 -> 124.1Improved
order fine-grained patches in Step then conv in Sequence (P+C vs C+P) (ablation)StARformer · Atari Breakout (offline RL)episodic return: 46.3 -> 124.1Improved
replace conv features with ViT patch embeddings in DT (ablation)DT · Atari Breakout (offline RL)episodic return: 70.7 -> 47.3Hurt
layer-wise attention over all tokens instead of summation (StAR Fusion) (ablation)StARformer · Atari Breakout (offline RL)episodic return: 29.2 -> 124.1Improved

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