A Robust Exploration Strategy in Reinforcement Learning Based on Temporal Difference Error.
Muhammad Shadi HajarHarsha K. KalutarageM. Omar Al-KadriPublished in: AI (2022)
Keyphrases
- temporal difference
- reinforcement learning
- exploration strategy
- function approximation
- td learning
- reinforcement learning algorithms
- evaluation function
- model free
- policy evaluation
- function approximators
- monte carlo
- action selection
- state space
- policy iteration
- machine learning
- markov decision processes
- markov chain
- markov decision problems
- action space
- optimal policy
- supervised learning
- partially observable
- training set
- multi agent