Discovery of pattern meaning from delayed rewards by reinforcement learning with a recurrent neural network.
Katsunari ShibataHiroki UtsunomiyaPublished in: IJCNN (2011)
Keyphrases
- reinforcement learning
- pattern discovery
- function approximation
- markov decision processes
- reinforcement learning algorithms
- state space
- reward shaping
- learning algorithm
- pattern matching
- machine learning
- discovery process
- reward function
- temporal difference
- knowledge discovery
- natural language
- optimal policy
- dynamic programming
- optimal control
- model free
- multi agent
- pattern detection
- data mining
- transfer learning
- pattern mining
- action selection
- data analysis
- multi agent systems