Extending immediate reinforcement learning on neural networks to multiple actions.
Claude F. TouzetPublished in: ESANN (1994)
Keyphrases
- reinforcement learning
- neural network
- artificial neural networks
- fuzzy logic
- recurrent neural networks
- action space
- plan recognition
- decision theoretic
- partially observable domains
- perceptual aliasing
- multiagent reinforcement learning
- situation calculus
- genetic algorithm
- model free
- action selection
- function approximation
- policy search
- back propagation
- state and action spaces
- dynamic programming