Linear regression MDP scheme for discrete backward stochastic differential equations under general conditions.
Emmanuel GobetPlamen TurkedjievPublished in: Math. Comput. (2016)
Keyphrases
- linear regression
- general conditions
- stochastic differential equations
- maximum a posteriori estimation
- least squares
- brownian motion
- additive gaussian noise
- markov decision processes
- vector valued
- reinforcement learning
- fractional brownian motion
- optimal policy
- image denoising
- multiscale
- object recognition
- non stationary
- video sequences
- differential equations
- viewing conditions