Bi-Lipschitz Ansatz for Anti-Symmetric Functions

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Dym, Nadav, Lu, Jianfeng, Mizrachi, Matan
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915184140877824
author Dym, Nadav
Lu, Jianfeng
Mizrachi, Matan
author_facet Dym, Nadav
Lu, Jianfeng
Mizrachi, Matan
contents Motivated by applications for simulating quantum many body functions, we propose a new universal ansatz for approximating anti-symmetric functions. The main advantage of this ansatz over previous alternatives is that it is bi-Lipschitz with respect to a naturally defined metric. As a result, we are able to obtain quantitative approximation results for approximation of Lipschitz continuous antisymmetric functions. Moreover, we provide preliminary experimental evidence to the improved performance of this ansatz for learning antisymmetric functions.
format Preprint
id arxiv_https___arxiv_org_abs_2503_04263
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Bi-Lipschitz Ansatz for Anti-Symmetric Functions
Dym, Nadav
Lu, Jianfeng
Mizrachi, Matan
Machine Learning
Classical Analysis and ODEs
Quantum Physics
41A63, 41A29, 68T07, 81Q05
G.1.2; I.5.1; G.1.10; I.5.4
Motivated by applications for simulating quantum many body functions, we propose a new universal ansatz for approximating anti-symmetric functions. The main advantage of this ansatz over previous alternatives is that it is bi-Lipschitz with respect to a naturally defined metric. As a result, we are able to obtain quantitative approximation results for approximation of Lipschitz continuous antisymmetric functions. Moreover, we provide preliminary experimental evidence to the improved performance of this ansatz for learning antisymmetric functions.
title Bi-Lipschitz Ansatz for Anti-Symmetric Functions
topic Machine Learning
Classical Analysis and ODEs
Quantum Physics
41A63, 41A29, 68T07, 81Q05
G.1.2; I.5.1; G.1.10; I.5.4
url https://arxiv.org/abs/2503.04263