A data-driven and model-based accelerated Hamiltonian Monte Carlo method for Bayesian elliptic inverse problems.
Sijing LiCheng ZhangZhiwen ZhangHongkai ZhaoPublished in: CoRR (2021)
Keyphrases
- data driven
- inverse problems
- monte carlo method
- markov chain
- posterior distribution
- partial differential equations
- bayesian learning
- image reconstruction
- monte carlo
- global optimization
- convex optimization
- maximum a posteriori
- optimization problems
- optimization methods
- genetic algorithm
- early vision
- probability distribution
- bayesian framework
- model selection
- maximum likelihood estimation
- image denoising
- posterior probability
- machine learning
- image processing
- optical flow
- level set
- high dimensional
- neural network
- evolutionary algorithm
- computer vision
- smoothness constraint
- super resolution