Bayesian deep learning for partial differential equation parameter discovery with sparse and noisy data.
Christophe BonnevilleChristopher J. EarlsPublished in: J. Comput. Phys. X (2022)
Keyphrases
- noisy data
- deep learning
- partial differential equations
- high dimensional
- image denoising
- unsupervised learning
- level set
- anisotropic diffusion
- image processing
- machine learning
- missing data
- multiscale
- high order
- image enhancement
- input data
- bayesian networks
- missing values
- data mining
- maximum likelihood
- supervised learning
- decision trees
- mental models
- weakly supervised
- shape prior
- sparse representation
- image segmentation
- image restoration
- training data
- dimensionality reduction
- higher order
- image data
- pairwise
- feature space