Texonom
Texonom/
Engineering
Engineering/Data Engineering/Artificial Intelligence/Machine Learning/Machine Learning Theory/Reinforcement Learning/TRL/
Decision Transformer
Search

Decision Transformer

Creator
Creator
Seonglae ChoSeonglae Cho
Created
Created
2024 Mar 8 5:53
Editor
Editor
Seonglae ChoSeonglae Cho
Edited
Edited
2024 May 31 3:33
Refs
Refs
Tokenizes states, actions, and rewards but emphasizes accurately predicting future actions from these sequences
  • Return conditioned policy can be used for policy rollout
  1. desired return
  1. subtract reward
notion image
 
 
Works well for long horizon and sparse task compared to traditional RL
Does success of Mar false approach
  • More supervision and MDP can be differ per task
 
 
 
 
 
 
NeuralPS 2021
Decision Transformer: Reinforcement Learning via Sequence Modeling
We introduce a framework that abstracts Reinforcement Learning (RL) as a sequence modeling problem. This allows us to draw upon the simplicity and scalability of the Transformer architecture, and associated advances in…
Decision Transformer: Reinforcement Learning via Sequence Modeling
https://ar5iv.labs.arxiv.org/html/2106.01345
Decision Transformer: Reinforcement Learning via Sequence Modeling
 
 

Recommendations

Texonom
Texonom/
Engineering
Engineering/Data Engineering/Artificial Intelligence/Machine Learning/Machine Learning Theory/Reinforcement Learning/TRL/
Decision Transformer