Estimating hyperparameters and instrument parameters in regularized inversion. Illustration for SPIRE/Herschel map making
François OrieuxJean-François GiovannelliThomas RodetAlain AbergelPublished in: CoRR (2012)
Keyphrases
- hyperparameters
- maximum a posteriori
- model selection
- image reconstruction
- cross validation
- maximum likelihood
- closed form
- bayesian inference
- support vector
- bayesian framework
- gaussian process
- markov random field
- posterior distribution
- prior information
- random sampling
- em algorithm
- bayesian methods
- noise level
- incremental learning
- expectation maximization
- regularization parameter
- parameter optimization
- parameter settings
- sample size
- parameter space
- incomplete data
- marginal likelihood
- missing values
- parameter estimation
- gaussian processes
- energy function
- objective function
- data sets
- generative model
- super resolution
- high resolution
- pairwise
- data mining