Parsimonious representation of nonlinear dynamical systems through manifold learning: a chemotaxis case study
From MaRDI portal
Abstract: Nonlinear manifold learning algorithms, such as diffusion maps, have been fruitfully applied in recent years to the analysis of large and complex data sets. However, such algorithms still encounter challenges when faced with real data. One such challenge is the existence of "repeated eigendirections," which obscures the detection of the true dimensionality of the underlying manifold and arises when several embedding coordinates parametrize the same direction in the intrinsic geometry of the data set. We propose an algorithm, based on local linear regression, to automatically detect coordinates corresponding to repeated eigendirections. We construct a more parsimonious embedding using only the eigenvectors corresponding to unique eigendirections, and we show that this reduced diffusion maps embedding induces a metric which is equivalent to the standard diffusion distance. We first demonstrate the utility and flexibility of our approach on synthetic data sets. We then apply our algorithm to data collected from a stochastic model of cellular chemotaxis, where our approach for factoring out repeated eigendirections allows us to detect changes in dynamical behavior and the underlying intrinsic system dimensionality directly from data.
Recommendations
- Dimensionality reduction of collective motion by principal manifolds
- Manifold learning for data-driven dynamical system analysis
- Principal Manifolds and Nonlinear Dimensionality Reduction via Tangent Space Alignment
- Landmark diffusion maps (L-dMaps): accelerated manifold learning out-of-sample extension
- Nonlinear Dimension Reduction by Local Multidimensional Scaling
Cites work
- All of Nonparametric Statistics
- Diffusion maps, spectral clustering and reaction coordinates of dynamical systems
- Geometric diffusions as a tool for harmonic analysis and structure definition of data: diffusion maps
- Geometric harmonics: a novel tool for multiscale out-of-sample extension of empirical functions
- Graph Laplacian Tomography From Unknown Random Projections
- Group invariant scattering
- Hölder-Lipschitz norms and their duals on spaces with semigroups, with applications to earth mover's distance
- Image Completion by Diffusion Maps and Spectral Relaxation
- Intrinsic modeling of stochastic dynamical systems using empirical geometry
- Laplacian Eigenmaps for Dimensionality Reduction and Data Representation
- Manifold Learning for Latent Variable Inference in Dynamical Systems
- Manifold parametrizations by eigenfunctions of the Laplacian and heat kernels
- Models of dispersal in biological systems
- Non-linear independent component analysis with diffusion maps
- The Diffusion Limit of Transport Equations Derived from Velocity-Jump Processes
- The earth mover's distance as a metric for image retrieval
- Time-scale separation from diffusion-mapped delay coordinates
- Viewing angle classification of cryo-electron microscopy images using eigenvectors
Cited in
(28)- Dimensionality reduction of complex metastable systems via kernel embeddings of transition manifolds
- Equation-free patch scheme for efficient computational homogenisation via self-adjoint coupling
- Numerical bifurcation analysis of PDEs from lattice Boltzmann model simulations: a parsimonious machine learning approach
- A geometric heat-flow theory of Lagrangian coherent structures
- Manifold learning for accelerating coarse-grained optimization
- Manifold learning for data-driven dynamical system analysis
- Data-driven control of agent-based models: an equation/variable-free machine learning approach
- Dimensionality reduction of collective motion by principal manifolds
- Data-Driven Reduction for a Class of Multiscale Fast-Slow Stochastic Dynamical Systems
- Manifold learning for organizing unstructured sets of process observations
- Kernel-based parameter estimation of dynamical systems with unknown observation functions
- Deep learning models for global coordinate transformations that linearise PDEs
- scientific article; zbMATH DE number 7626797 (Why is no real title available?)
- Spectral Discovery of Jointly Smooth Features for Multimodal Data
- Spectral embedding norm: looking deep into the spectrum of the graph Laplacian
- Grassmannian diffusion maps based surrogate modeling via geometric harmonics
- Physics-agnostic and physics-infused machine learning for thin films flows: modelling, and predictions from small data
- Learning black- and gray-box chemotactic PDEs/closures from agent based Monte Carlo simulation data
- Double diffusion maps and their latent harmonics for scientific computations in latent space
- Initializing LSTM internal states via manifold learning
- Time-series forecasting using manifold learning, radial basis function interpolation, and geometric harmonics
- Learning effective stochastic differential equations from microscopic simulations: linking stochastic numerics to deep learning
- Tipping points of evolving epidemiological networks: machine learning-assisted, data-driven effective modeling
- A resolution independent neural operator
- GoRINNs: Godunov-Riemann informed neural networks for learning hyperbolic conservation laws
- Enabling probabilistic learning on manifolds through double diffusion maps
- PDE-free mass-constrained learning of complex systems with hidden states
- Autoencoders vs. Numerical Analysis--Informed Manifold Learning for Navier--Stokes Flows
This page was built for publication: Parsimonious representation of nonlinear dynamical systems through manifold learning: a chemotaxis case study
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1742825)
