Sparse Temporal Difference Learning via Alternating Direction Method of Multipliers.
Nikos TsipinakisJames D. B. NelsonPublished in: ICMLA (2015)
Keyphrases
- temporal difference learning
- alternating direction method of multipliers
- function approximation
- fixed point
- basis pursuit
- convex optimization
- evaluation function
- reinforcement learning
- game playing
- temporal difference
- total variation
- markov decision process
- compressed sensing
- sparse representation
- denoising
- gaussian process
- reinforcement learning algorithms
- high dimensional
- monte carlo
- learning algorithm
- policy iteration
- function approximators
- matrix completion
- sufficient conditions
- image classification