Efficient Reinforcement Learning in Block MDPs: A Model-free Representation Learning Approach.
Xuezhou ZhangYuda SongMasatoshi UeharaMengdi WangAlekh AgarwalWen SunPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- model free
- reinforcement learning algorithms
- markov decision processes
- function approximation
- policy iteration
- learning problems
- state space
- learning algorithm
- hierarchical reinforcement learning
- rl algorithms
- learning process
- reinforcement learning methods
- partially observable
- supervised learning
- temporal difference learning
- dynamic programming
- policy evaluation
- continuous state
- multi agent
- machine learning
- temporal difference
- model based reinforcement learning
- average reward
- transfer learning
- neural network
- continuous state and action spaces
- markov decision problems
- learning agent
- action selection
- support vector machine svm