An Empirical Comparison of Two Common Multiobjective Reinforcement Learning Algorithms.
Rustam IssabekovPeter VamplewPublished in: Australasian Conference on Artificial Intelligence (2012)
Keyphrases
- reinforcement learning algorithms
- multi objective
- reinforcement learning
- state space
- evolutionary algorithm
- markov decision processes
- model free
- multiobjective optimization
- optimization algorithm
- multi objective optimization
- reinforcement learning methods
- eligibility traces
- genetic algorithm
- reinforcement learning problems
- pareto optimal
- temporal difference
- learning algorithm
- objective function
- dynamic environments
- particle swarm optimization
- multiple objectives
- reward function
- partially observable environments
- nsga ii
- function approximation
- policy search
- multiagent reinforcement learning
- sufficient conditions
- reward shaping
- mobile robot