A co-evolutionary approach to interpretable reinforcement learning in environments with continuous action spaces.
Leonardo Lucio CustodeGiovanni IaccaPublished in: SSCI (2021)
Keyphrases
- action space
- reinforcement learning
- state space
- markov decision processes
- real valued
- state and action spaces
- continuous state
- continuous state spaces
- action selection
- reinforcement learning methods
- skill learning
- stochastic processes
- markov decision process
- control policies
- continuous action
- state action
- function approximation
- optimal policy
- machine learning
- dynamic programming
- single agent
- markov decision problems
- function approximators
- model free
- maximum entropy
- path planning
- heuristic search
- dynamic environments
- learning algorithm
- path finding
- reinforcement learning algorithms
- finite state
- autonomous robots
- state variables
- dynamical systems
- reinforcement learning problems
- multi agent