Modulating reinforcement-learning parameters using agent emotions.
Rickard von HaugwitzYoshifumi KitamuraKazuki TakashimaPublished in: SCIS&ISIS (2012)
Keyphrases
- reinforcement learning
- multi agent
- learning agent
- action selection
- parameter values
- autonomous agents
- multiagent systems
- intelligent agents
- multi agent systems
- dynamic environments
- machine learning
- learning agents
- markov decision process
- learning capabilities
- decision making
- parameter estimation
- neural network
- partially observable
- dynamic programming
- function approximation
- maximum likelihood
- parameter space
- model free
- state space
- temporal difference
- agent systems
- learning algorithm
- virtual agents
- mobile agents
- autonomous learning
- multi agent environments