A supervised learning scheme for computing Hamilton-Jacobi equation via density coupling (Q6876882)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 8143003
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | A supervised learning scheme for computing Hamilton-Jacobi equation via density coupling |
scientific article; zbMATH DE number 8143003 |
Statements
A supervised learning scheme for computing Hamilton-Jacobi equation via density coupling (English)
0 references
9 January 2026
0 references
The authors discuss a numerical method to solve (forward) first-order Hamilton-Jacobi equation HJ with a convex Hamiltonian \(H\) based on its geometric interpretation via optimal transportation. The particle formulation of Wasserstein flow is solved designing the supervised learning scheme. Here, the loss function is taken as the Bregman divergence of \(H\). This scheme can be used to solve HJ even after appearance of singularities (caustics).
0 references
Hamilton-Jacobi
0 references
Wasserstein flow
0 references
density coupling
0 references
supervised learning
0 references
0 references
0 references
0 references
0 references
0 references
0 references
0 references
0 references
0 references
0 references
