Sample-efficient Learning of Infinite-horizon Average-reward MDPs with General Function Approximation.
Jianliang HeHan ZhongZhuoran YangPublished in: CoRR (2024)
Keyphrases
- function approximation
- average reward
- markov decision processes
- infinite horizon
- reinforcement learning
- efficient learning
- optimal policy
- policy iteration
- long run
- finite horizon
- model free
- reinforcement learning algorithms
- actor critic
- stochastic games
- state space
- temporal difference
- total reward
- finite state
- average cost
- dynamic programming
- markov decision process
- policy evaluation
- learning algorithm
- discount factor
- partially observable
- td learning
- discounted reward
- policy gradient
- markov decision problems
- optimal control
- learning tasks
- function approximators
- decision problems
- partially observable markov decision processes
- reward function
- dec pomdps
- initial state
- lead time
- fixed point
- machine learning
- data mining
- state action
- continuous state