DL-PDE: Deep-learning based data-driven discovery of partial differential equations from discrete and noisy data.
Hao XuHaibin ChangDongxiao ZhangPublished in: CoRR (2019)
Keyphrases
- partial differential equations
- noisy data
- deep learning
- difference equations
- level set
- image denoising
- anisotropic diffusion
- image processing
- numerical solution
- unsupervised learning
- machine learning
- diffusion equation
- multiscale
- numerical methods
- conservation laws
- energy functional
- missing data
- image enhancement
- eikonal equation
- differential equations
- high dimensionality
- high dimensional
- weakly supervised
- curve evolution
- numerical algorithms
- mental models
- fourth order
- denoising
- natural images
- training data
- finite difference
- active contours
- high order
- knn
- object detection
- pattern recognition
- computer vision
- data sets