Rethinking the Discount Factor in Reinforcement Learning: A Decision Theoretic Approach.
Silviu PitisPublished in: AAAI (2019)
Keyphrases
- np hard
- discount factor
- reinforcement learning
- optimal policy
- markov decision processes
- partially observable
- markov decision problems
- average reward
- decision problems
- state space
- reinforcement learning algorithms
- dynamic programming
- policy iteration
- function approximation
- finite state
- infinite horizon
- action space
- learning algorithm
- model free
- multi agent
- temporal difference
- single agent
- optimal control
- long run
- reward function
- sufficient conditions
- learning process
- initial state
- machine learning
- decision processes
- computational complexity
- expected utility
- action selection
- transfer learning
- neural network