KaBaGe-RL: Kanerva-based generalisation and reinforcement learning for possession football.
Kostas KostiadisHuosheng HuPublished in: IROS (2001)
Keyphrases
- function approximation
- reinforcement learning
- model free
- function approximators
- temporal difference
- state space
- temporal difference learning
- reinforcement learning algorithms
- radial basis function
- rl algorithms
- learning algorithm
- dynamic programming
- optimal policy
- markov decision processes
- receptive fields
- learning problems
- machine learning
- learning classifier systems
- multi agent
- optimal control
- supervised learning
- learning process
- reinforcement learning methods
- neural network
- policy gradient
- learning tasks
- transfer learning
- linear combination
- markov decision process
- artificial neural networks
- actor critic
- partially observable domains
- direct policy search