Convergence in Navigational Reinforcement Learning.
Tom J. AmelootJan Van den BusschePublished in: CoRR (2015)
Keyphrases
- reinforcement learning
- stochastic approximation
- function approximation
- model free
- state space
- supervised learning
- machine learning
- reinforcement learning algorithms
- action selection
- convergence rate
- learning algorithm
- temporal difference
- global convergence
- sufficient conditions
- optimal policy
- markov decision processes
- learning problems
- learning process
- objective function
- decision making
- genetic algorithm
- learning capabilities
- convergence analysis
- neural network