Neuroevolutionary diversity policy search for multi-objective reinforcement learning.
Dan ZhouJiqing DuSachiyo AraiPublished in: Inf. Sci. (2024)
Keyphrases
- policy search
- multi objective
- reinforcement learning
- function approximators
- temporal difference
- reinforcement learning algorithms
- evolutionary algorithm
- multi objective optimization
- function approximation
- continuous state
- genetic algorithm
- objective function
- particle swarm optimization
- multiple objectives
- continuous action
- state space
- dynamic programming
- reward function
- policy gradient
- neural network
- optimal policy
- nsga ii
- machine learning
- markov decision processes
- multi agent
- model free
- policy iteration
- approximation methods
- reinforcement learning methods
- markov decision problems
- optimal control
- weight vector
- evaluation function
- partially observable markov decision processes
- real valued
- control policies
- hidden state
- learning tasks
- supervised learning