Trajectory-Aware Eligibility Traces for Off-Policy Reinforcement Learning.
Brett DaleyMartha WhiteChristopher AmatoMarlos C. MachadoPublished in: ICML (2023)
Keyphrases
- eligibility traces
- reinforcement learning
- reinforcement learning algorithms
- reinforcement learning methods
- policy evaluation
- function approximation
- state space
- model free
- temporal difference
- machine learning
- multi agent
- control problems
- markov decision processes
- learning speed
- optimal policy
- transfer learning
- markov decision process
- function approximators
- learning process
- search space