Tensor and Matrix Low-Rank Value-Function Approximation in Reinforcement Learning.
Sergio RozadaSantiago PaternainAntonio G. MarquesPublished in: IEEE Trans. Signal Process. (2024)
Keyphrases
- low rank
- trace norm
- reinforcement learning
- frobenius norm
- high order
- temporal difference
- state space
- tensor decomposition
- convex optimization
- matrix completion
- linear combination
- missing data
- rank minimization
- matrix factorization
- low rank matrix
- singular value decomposition
- matrix decomposition
- function approximation
- high dimensional data
- semi supervised
- singular values
- basis functions
- factorization methods
- data matrix
- kernel matrix
- low rank matrices
- function approximators
- reinforcement learning algorithms
- low rank and sparse
- robust principal component analysis
- affinity matrix
- image processing
- machine learning
- tensor factorization
- markov decision process
- model free
- multi task
- markov decision processes
- higher order
- small number
- nuclear norm
- low rank approximation
- interior point methods
- norm minimization
- learning algorithm
- data sets