SESaMo: Symmetry-Enforcing Stochastic Modulation for Normalizing Flows

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kreit, Janik, Schuh, Dominic, Nicoli, Kim A., Funcke, Lena
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908395834966016
author Kreit, Janik
Schuh, Dominic
Nicoli, Kim A.
Funcke, Lena
author_facet Kreit, Janik
Schuh, Dominic
Nicoli, Kim A.
Funcke, Lena
contents Deep generative models have recently garnered significant attention across various fields, from physics to chemistry, where sampling from unnormalized Boltzmann-like distributions represents a fundamental challenge. In particular, autoregressive models and normalizing flows have become prominent due to their appealing ability to yield closed-form probability densities. Moreover, it is well-established that incorporating prior knowledge - such as symmetries - into deep neural networks can substantially improve training performances. In this context, recent advances have focused on developing symmetry-equivariant generative models, achieving remarkable results. Building upon these foundations, this paper introduces Symmetry-Enforcing Stochastic Modulation (SESaMo). Similar to equivariant normalizing flows, SESaMo enables the incorporation of inductive biases (e.g., symmetries) into normalizing flows through a novel technique called stochastic modulation. This approach enhances the flexibility of the generative model, allowing to effectively learn a variety of exact and broken symmetries. Our numerical experiments benchmark SESaMo in different scenarios, including an 8-Gaussian mixture model and physically relevant field theories, such as the $ϕ^4$ theory and the Hubbard model.
format Preprint
id arxiv_https___arxiv_org_abs_2505_19619
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SESaMo: Symmetry-Enforcing Stochastic Modulation for Normalizing Flows
Kreit, Janik
Schuh, Dominic
Nicoli, Kim A.
Funcke, Lena
Machine Learning
Strongly Correlated Electrons
High Energy Physics - Lattice
Computational Physics
Deep generative models have recently garnered significant attention across various fields, from physics to chemistry, where sampling from unnormalized Boltzmann-like distributions represents a fundamental challenge. In particular, autoregressive models and normalizing flows have become prominent due to their appealing ability to yield closed-form probability densities. Moreover, it is well-established that incorporating prior knowledge - such as symmetries - into deep neural networks can substantially improve training performances. In this context, recent advances have focused on developing symmetry-equivariant generative models, achieving remarkable results. Building upon these foundations, this paper introduces Symmetry-Enforcing Stochastic Modulation (SESaMo). Similar to equivariant normalizing flows, SESaMo enables the incorporation of inductive biases (e.g., symmetries) into normalizing flows through a novel technique called stochastic modulation. This approach enhances the flexibility of the generative model, allowing to effectively learn a variety of exact and broken symmetries. Our numerical experiments benchmark SESaMo in different scenarios, including an 8-Gaussian mixture model and physically relevant field theories, such as the $ϕ^4$ theory and the Hubbard model.
title SESaMo: Symmetry-Enforcing Stochastic Modulation for Normalizing Flows
topic Machine Learning
Strongly Correlated Electrons
High Energy Physics - Lattice
Computational Physics
url https://arxiv.org/abs/2505.19619