Discovery of partial differential equations from highly noisy and sparse data with physics-informed information criterion.
Hao XuJunsheng ZengDongxiao ZhangPublished in: CoRR (2022)
Keyphrases
- sparse data
- partial differential equations
- information criterion
- cross validation
- model selection
- high dimensional
- probability model
- image processing
- image denoising
- level set
- image enhancement
- multiscale
- bayesian information criterion
- widely applicable
- data mining
- factor analysis
- sample size
- data cube
- random projections
- graph cuts
- high order
- data analysis
- variable selection
- pattern recognition