A Temporal Difference GNG-Based Approach for the State Space Quantization in Reinforcement Learning Environments.
Davi Carnauba de Lima VieiraPaulo Jorge Leitão AdeodatoPaulo M. GoncalvesPublished in: ICTAI (2013)
Keyphrases
- reinforcement learning
- temporal difference
- state space
- learning environment
- reinforcement learning algorithms
- function approximation
- td learning
- model free
- heuristic search
- learning process
- optimal policy
- growing neural gas
- action selection
- markov decision processes
- policy evaluation
- markov chain
- policy iteration
- machine learning
- temporal difference methods
- function approximators
- planning problems
- supervised learning
- partially observable
- dynamic programming
- state variables
- initial state
- particle filter
- learning agent
- transfer learning
- e learning
- learning algorithm
- neural network
- optimal control
- dynamical systems
- action space