Low-rank State-action Value-function Approximation.
Sergio RozadaVictor TenorioAntonio G. MarquesPublished in: EUSIPCO (2021)
Keyphrases
- state action
- low rank
- kernel matrix
- evaluation function
- reinforcement learning
- linear combination
- missing data
- matrix factorization
- semidefinite programming
- convex optimization
- singular value decomposition
- semi supervised
- high dimensional data
- stochastic games
- markov decision process
- average reward
- high order
- state transitions
- data sets
- supervised learning
- state space
- reward function
- action space
- small number
- markov decision processes
- basis functions
- image restoration