Using Localization and Factorization to Reduce the Complexity of Reinforcement Learning.
Peter SunehagMarcus HutterPublished in: AGI (2015)
Keyphrases
- reinforcement learning
- function approximation
- reinforcement learning algorithms
- matrix factorization
- learning algorithm
- state space
- learning process
- robotic control
- markov decision processes
- low rank
- pairwise
- localization method
- optimal policy
- temporal difference
- kronecker product
- machine learning
- learning problems
- model free
- localization error
- function approximators
- multi agent reinforcement learning
- stochastic approximation
- genetic algorithm
- multi agent
- non rigid structure from motion
- source localization
- dynamic programming
- optic disc
- localization algorithm
- factorization method
- position information
- multibody
- robot control
- learning capabilities
- missing data