LSTD with Random Projections.
Mohammad GhavamzadehAlessandro LazaricOdalric-Ambrym MaillardRémi MunosPublished in: NIPS (2010)
Keyphrases
- random projections
- reinforcement learning
- temporal difference
- policy evaluation
- least squares
- policy iteration
- function approximation
- dimensionality reduction
- model free
- monte carlo
- markov decision processes
- image reconstruction
- dimension reduction
- sparse representation
- original data
- evaluation function
- random sampling
- optimal policy
- action selection
- state space
- principal component analysis
- variance reduction
- reinforcement learning algorithms
- machine learning
- low dimensional
- high dimensionality
- document clustering
- computer vision
- data sets
- dynamic programming