Solution Theory of Hamilton-Jacobi-Bellman Equations in Spectral Barron Spaces

Ye Feng, Jianfeng Lu · 2025

Read Solution Theory of Hamilton-Jacobi-Bellman Equations in Spectral Barron Spaces by Ye Feng, Jianfeng Lu (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-18656.txt.

Description

Ye Feng and Jianfeng Lu develop solution theory for a class of control equations in a 2025 paper. Hamilton-Jacobi-Bellman equations govern optimal choices in systems that evolve over time, and they are notoriously hard in high dimensions. The authors work in spectral Barron spaces, a function class tied to how shallow neural networks represent functions.

They assume the equation's coefficients live in that class and the discount factor is large enough. Under those conditions they build a sequence of bounded functions of the same class that converges to the true solution, locally and evenly. As a result, two-layer neural networks can approximate the solution without the usual blow-up in cost as dimension grows. The result supports network-based solvers for high-dimensional control problems.

Catalog record

CREATORYe Feng, Jianfeng Lu
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE71 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-18656

Browse the full library · About the Tower of Babel