Continuous-domain assignment flows
From MaRDI portal
Abstract: Assignment flows denote a class of dynamical models for contextual data labeling (classification) on graphs. We derive a novel parametrization of assignment flows that reveals how the underlying information geometry induces two processes for assignment regularization and for gradually enforcing unambiguous decisions, respectively, that seamlessly interact when solving for the flow. Our result enables to characterize the dominant part of the assignment flow as a Riemannian gradient flow with respect to the underlying information geometry. We consider a continuous-domain formulation of the corresponding potential and develop a novel algorithm in terms of solving a sequence of linear elliptic PDEs subject to a simple convex constraint. Our result provides a basis for addressing learning problems by controlling such PDEs in future work.
Recommendations
Cites work
- A Convex Approach to Minimal Partitions
- A mean-field optimal control formulation of deep learning
- A proposal on machine learning via dynamical systems
- Approximation of functional depending on jumps by elliptic functional via t-convergence
- Assignment flows
- Continuous multiclass labeling approaches and algorithms
- Evolutionary game dynamics
- Geometric numerical integration of the assignment flow
- scientific article; zbMATH DE number 5869530 (Why is no real title available?)
- scientific article; zbMATH DE number 3933858 (Why is no real title available?)
- scientific article; zbMATH DE number 1560711 (Why is no real title available?)
- Image labeling by assignment
- Images as embedded maps and minimal surfaces: Movies, color, texture, and volumetric medical images
- Large data limit for a phase transition model with the p-Laplacian on point clouds
- Mirror descent and nonlinear projected subgradient methods for convex optimization.
- On the reaction-diffusion replicator systems: spatial patterns and asymptotic behaviour
- Optimal approximations by piecewise smooth functions and associated variational problems
- Proximal alternating linearized minimization for nonconvex and nonsmooth problems
- Replicator equations and space
- Riemannian geometry and geometric analysis
- Stable architectures for deep neural networks
- Unsupervised assignment flow: label learning on feature manifolds by spatially regularized geometric assignment
- Variational Analysis
- Variational Analysis in Sobolev and BV Spaces
- Vector-valued local minimizers of nonconvex variational problems
- Weakly Differentiable Functions
Cited in
(9)- Assignment flows for data labeling on graphs: convergence and stability
- Assignment flows
- Connections between deep learning and partial differential equations
- Self-assignment flows for unsupervised data labeling on graphs
- A Nonlocal Graph-PDE and Higher-Order Geometric Integration for Image Labeling
- On the geometric mechanics of assignment flows for metric data labeling
- Riemannian patch assignment gradient flows
- On the correspondence between replicator dynamics and assignment flows
- On the geometric mechanics of assignment flows for metric data labeling
This page was built for publication: Continuous-domain assignment flows
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5014843)
