Improving Reinforcement Learning through a Better Exploration Strategy and an Adjustable Representation of the Environment.
Roberto IglesiasMiguel A. RodríguezManuel SánchezEva PereiraCarlos V. RegueiroPublished in: EMCR (2007)
Keyphrases
- exploration strategy
- reinforcement learning
- unknown environments
- state space
- function approximation
- single agent
- markov decision processes
- temporal difference
- multi agent
- particle swarm optimisation
- locally optimal
- model free
- reinforcement learning algorithms
- machine learning
- supervised learning
- action selection
- action space
- optimal policy
- learning algorithm