Maximum Likelihood Estimation of Regularization Parameters in High-Dimensional Inverse Problems: An Empirical Bayesian Approach. Part II: Theoretical Analysis.
Valentin De BortoliAlain DurmusMarcelo PereyraAna Fernandez VidalPublished in: SIAM J. Imaging Sci. (2020)
Keyphrases
- maximum likelihood estimation
- inverse problems
- regularization method
- high dimensional
- regularization parameter
- image reconstruction
- hyperparameters
- global optimization
- parameter space
- em algorithm
- maximum likelihood
- image restoration
- convex optimization
- parameter estimation
- early vision
- low dimensional
- expectation maximization
- optimization problems
- cross validation
- dimensionality reduction
- probability distribution
- model selection
- smoothness constraint
- partial differential equations
- optimization methods
- high dimensional data
- feature space
- density function
- super resolution
- high quality
- markov random field
- objective function
- total variation
- closed form
- evolutionary algorithm