Symmetry Learning for Function Approximation in Reinforcement Learning.
Anuj MahajanTheja TulabandhulaPublished in: CoRR (2017)
Keyphrases
- reinforcement learning
- function approximation
- learning tasks
- temporal difference learning
- function approximators
- learning process
- temporal difference methods
- tile coding
- td learning
- learning algorithm
- temporal difference
- model free
- learning problems
- supervised learning
- state space
- continuous state
- action selection
- temporal difference learning algorithms
- reinforcement learning methods
- neural network
- mountain car
- radial basis function
- policy search
- optimal policy
- dynamic programming
- machine learning
- reinforcement learning problems
- actor critic
- evaluation function
- partially observable
- reinforcement learning algorithms
- markov decision processes
- learning experience
- machine learning algorithms
- prior knowledge
- artificial neural networks
- pattern recognition
- data mining