Reward-Relevance-Filtered Linear Offline Reinforcement Learning.
Angela ZhouPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- state space
- function approximation
- markov decision processes
- learning algorithm
- reward shaping
- dynamic programming
- reinforcement learning algorithms
- function approximators
- model free
- linear systems
- optimal policy
- partially observable
- relevance feedback
- eligibility traces
- multi agent
- information retrieval
- reinforcement learning methods
- temporal difference
- machine learning
- learning classifier systems
- linear model
- real time
- test collection
- supervised learning
- least squares
- learning process
- robotic control
- state and action spaces
- policy evaluation
- average reward
- reward function
- optimal control