Reinforcement Learning Reward Functions for Unsupervised Learning.
Colin FyfePei Ling LaiPublished in: ISNN (1) (2007)
Keyphrases
- reward function
- unsupervised learning
- reinforcement learning
- reinforcement learning algorithms
- markov decision processes
- supervised learning
- state space
- policy search
- partially observable
- function approximation
- optimal policy
- inverse reinforcement learning
- transition model
- markov decision process
- semi supervised
- multi agent
- model free
- temporal difference
- dynamic programming
- machine learning
- learning algorithm
- multiple agents
- markov decision problems
- initially unknown
- state action
- transition probabilities
- dimensionality reduction
- feature selection
- function approximators
- learning agent
- optimal control
- markov chain