Kernel Embedding Based Variational Approach for Low-Dimensional Approximation of Dynamical Systems.
Wenchong TianHao WuPublished in: Comput. Methods Appl. Math. (2021)
Keyphrases
- dynamical systems
- low dimensional
- graph embedding
- laplacian eigenmaps
- embedding space
- nonlinear dimensionality reduction
- manifold learning
- high dimensional
- input space
- multidimensional scaling
- latent space
- vector space
- dimensionality reduction
- feature space
- high dimensional data
- dynamic systems
- differential equations
- euclidean space
- principal component analysis
- kernel pca
- state space
- phase space
- data points
- kernel function
- pairwise distances
- control theory
- nonlinear dynamical systems
- linear systems
- qualitative simulation
- lower dimensional
- difference equations
- locally linear embedding
- dynamical behavior
- geometric structure
- image processing
- reinforcement learning
- partially observable markov decision processes
- gaussian processes
- nonlinear dynamics
- markov chain
- support vector
- agent environment
- objective function
- feature extraction