Exploration Strategy based on Validity of Actions in Deep Reinforcement Learning.
Hyung-Suk YoonSang-Hyun LeeSeung-Woo SeoPublished in: IROS (2020)
Keyphrases
- exploration strategy
- reinforcement learning
- action selection
- unknown environments
- partially observable
- function approximation
- action space
- state space
- reward function
- learning agent
- machine learning
- reinforcement learning algorithms
- optimal policy
- locally optimal
- model free
- markov decision problems
- optimal control
- temporal difference
- dynamic programming
- single agent
- learning algorithm
- decision theoretic
- mobile robot
- genetic algorithm
- outdoor environments