TP-XCS: An XCS classifier system with fixed-length memory for reinforcement learning.
Tom PickeringTim KovacsPublished in: CEC (2015)
Keyphrases
- fixed length
- learning classifier systems
- reinforcement learning
- variable length
- feature vectors
- classifier systems
- rule discovery
- real valued
- sequential data
- evolutionary computation
- vector representation
- state space
- genetic algorithm
- memory requirements
- evolutionary algorithm
- memory usage
- bitstream
- multiresolution
- model free
- machine learning
- n gram
- feature extraction
- function approximation
- temporal difference
- optimal policy
- language model
- computational intelligence
- image data
- learning process
- natural language
- multi agent
- multiscale
- learning algorithm
- information retrieval