An auxiliary variable method for Langevin based MCMC algorithms.
Yosra MarnissiÉmilie ChouzenouxJean-Christophe PesquetAmel Benazza-BenyahiaPublished in: SSP (2016)
Keyphrases
- theoretical analysis
- recently developed
- markov chain monte carlo
- computationally efficient
- dynamic programming
- computational cost
- objective function
- parameter estimation
- high accuracy
- computational efficiency
- similarity measure
- segmentation method
- pairwise
- classification algorithm
- clustering method
- noisy data
- probabilistic model
- significant improvement
- learning algorithm
- continuous variables
- high computational complexity
- particle filter
- optimization problems
- least squares
- prior knowledge
- data structure
- bayesian networks