Analyzing and Bridging the Gap between Maximizing Total Reward and Discounted Reward in Deep Reinforcement Learning.
Shuyu YinFei WenPeilin LiuTao LuoPublished in: CoRR (2024)
Keyphrases
- average reward
- total reward
- reinforcement learning
- discounted reward
- markov decision processes
- optimal policy
- model free
- policy iteration
- long run
- state and action spaces
- reinforcement learning algorithms
- optimality criterion
- hierarchical reinforcement learning
- function approximation
- state space
- finite state
- markov chain
- action selection
- action space
- temporal difference
- multistage
- transfer learning
- decision problems
- multi agent
- function approximators
- learning algorithm
- reward function
- probability distribution
- markov decision process
- machine learning
- linear program
- dynamical systems