Waveflow: Boundary-conditioned normalizing flows applied to fermionic wave functions

An efficient and expressive wave function Ansatz is key to scalable solutions for complex many-body electronic structures. While Slater determinants are predominantly used for constructing antisymmetric electronic wave function Ansätze, this construction can result in limited expressiveness when the...

Full description

Saved in:
Bibliographic Details
Main Authors: Luca Thiede, Chong Sun, Alán Aspuru-Guzik
Format: Article
Language:English
Published: AIP Publishing LLC 2024-12-01
Series:APL Machine Learning
Online Access:http://dx.doi.org/10.1063/5.0229620
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850040067751411712
author Luca Thiede
Chong Sun
Alán Aspuru-Guzik
author_facet Luca Thiede
Chong Sun
Alán Aspuru-Guzik
author_sort Luca Thiede
collection DOAJ
description An efficient and expressive wave function Ansatz is key to scalable solutions for complex many-body electronic structures. While Slater determinants are predominantly used for constructing antisymmetric electronic wave function Ansätze, this construction can result in limited expressiveness when the targeted wave function is highly complex. In this work, we introduce Waveflow, an innovative framework for learning many-body fermionic wave functions using boundary-conditioned normalizing flows. Instead of relying on Slater determinants, Waveflow imposes antisymmetry by defining the fundamental domain of the wave function and applying necessary boundary conditions. A key challenge in using normalizing flows for this purpose is addressing the topological mismatch between the prior and target distributions. We propose using O-spline priors and I-spline bijections to handle this mismatch, which allows for flexibility in the node number of the distribution while automatically maintaining its square-normalization property. We apply Waveflow to a one-dimensional many-electron system, where we variationally minimize the system’s energy using variational quantum Monte Carlo (VQMC). Our experiments demonstrate that Waveflow can effectively resolve topological mismatches and faithfully learn the ground-state wave function.
format Article
id doaj-art-4e555d7a719045a99096a5d1d226e745
institution DOAJ
issn 2770-9019
language English
publishDate 2024-12-01
publisher AIP Publishing LLC
record_format Article
series APL Machine Learning
spelling doaj-art-4e555d7a719045a99096a5d1d226e7452025-08-20T02:56:11ZengAIP Publishing LLCAPL Machine Learning2770-90192024-12-0124046106046106-1210.1063/5.0229620Waveflow: Boundary-conditioned normalizing flows applied to fermionic wave functionsLuca Thiede0Chong Sun1Alán Aspuru-Guzik2Department of Computer Science, University of Toronto, Toronto, Ontario M5S 2E4, CanadaDepartment of Computer Science, University of Toronto, Toronto, Ontario M5S 2E4, CanadaDepartment of Computer Science, University of Toronto, Toronto, Ontario M5S 2E4, CanadaAn efficient and expressive wave function Ansatz is key to scalable solutions for complex many-body electronic structures. While Slater determinants are predominantly used for constructing antisymmetric electronic wave function Ansätze, this construction can result in limited expressiveness when the targeted wave function is highly complex. In this work, we introduce Waveflow, an innovative framework for learning many-body fermionic wave functions using boundary-conditioned normalizing flows. Instead of relying on Slater determinants, Waveflow imposes antisymmetry by defining the fundamental domain of the wave function and applying necessary boundary conditions. A key challenge in using normalizing flows for this purpose is addressing the topological mismatch between the prior and target distributions. We propose using O-spline priors and I-spline bijections to handle this mismatch, which allows for flexibility in the node number of the distribution while automatically maintaining its square-normalization property. We apply Waveflow to a one-dimensional many-electron system, where we variationally minimize the system’s energy using variational quantum Monte Carlo (VQMC). Our experiments demonstrate that Waveflow can effectively resolve topological mismatches and faithfully learn the ground-state wave function.http://dx.doi.org/10.1063/5.0229620
spellingShingle Luca Thiede
Chong Sun
Alán Aspuru-Guzik
Waveflow: Boundary-conditioned normalizing flows applied to fermionic wave functions
APL Machine Learning
title Waveflow: Boundary-conditioned normalizing flows applied to fermionic wave functions
title_full Waveflow: Boundary-conditioned normalizing flows applied to fermionic wave functions
title_fullStr Waveflow: Boundary-conditioned normalizing flows applied to fermionic wave functions
title_full_unstemmed Waveflow: Boundary-conditioned normalizing flows applied to fermionic wave functions
title_short Waveflow: Boundary-conditioned normalizing flows applied to fermionic wave functions
title_sort waveflow boundary conditioned normalizing flows applied to fermionic wave functions
url http://dx.doi.org/10.1063/5.0229620
work_keys_str_mv AT lucathiede waveflowboundaryconditionednormalizingflowsappliedtofermionicwavefunctions
AT chongsun waveflowboundaryconditionednormalizingflowsappliedtofermionicwavefunctions
AT alanaspuruguzik waveflowboundaryconditionednormalizingflowsappliedtofermionicwavefunctions