Convolutional Neural Networks for Very Low-Dimensional LPV Approximations of Incompressible Navier-Stokes Equations.
Jan HeilandPeter BennerRezvan BahmaniPublished in: Frontiers Appl. Math. Stat. (2022)
Keyphrases
- convolutional neural networks
- low dimensional
- navier stokes equations
- high dimensional
- convolutional network
- dimensionality reduction
- high dimensional data
- boundary conditions
- manifold learning
- principal component analysis
- data points
- feature space
- linear dimensionality reduction
- multidimensional scaling
- deformable models
- euclidean space
- velocity field
- input space
- nonlinear dimensionality reduction
- takagi sugeno
- closed form
- linear subspace
- graph embedding
- neural network
- pairwise
- image processing