Learning Infinite-horizon Average-reward MDPs with Linear Function Approximation.
Chen-Yu WeiMehdi Jafarnia-JahromiHaipeng LuoRahul JainPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- function approximation
- markov decision processes
- average reward
- infinite horizon
- optimal policy
- policy iteration
- actor critic
- td learning
- stochastic games
- finite horizon
- state space
- learning tasks
- partially observable
- reinforcement learning algorithms
- function approximators
- optimal control
- model free
- total reward
- long run
- temporal difference
- dynamic programming
- markov decision problems
- policy gradient
- learning algorithm
- reinforcement learning methods
- rl algorithms
- discount factor
- markov decision process
- average cost
- action selection
- finite state
- evaluation function
- supervised learning
- machine learning
- partially observable markov decision processes
- graphical models
- policy evaluation
- multi agent
- data mining