Transformers Learn Temporal Difference Methods for In-Context Reinforcement Learning.
Jiuqi WangEthan BlaserHadi DaneshmandShangtong ZhangPublished in: CoRR (2024)
Keyphrases
- temporal difference methods
- function approximators
- reinforcement learning
- function approximation
- temporal difference
- policy search
- machine learning
- learning algorithm
- reinforcement learning problems
- learning agent
- least squares
- learning process
- transfer learning
- learning tasks
- markov chain
- supervised learning
- dynamic programming
- cost function
- multi agent