How to Square Tensor Networks and Circuits Without Squaring Them

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Loconte, Lorenzo, Javaloy, Adrián, Vergari, Antonio
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866914603955388416
author Loconte, Lorenzo
Javaloy, Adrián
Vergari, Antonio
author_facet Loconte, Lorenzo
Javaloy, Adrián
Vergari, Antonio
contents Squared tensor networks (TNs) and their extension as computational graphs--squared circuits--have been used as expressive distribution estimators, yet supporting closed-form marginalization. However, the squaring operation introduces additional complexity when computing the partition function or marginalizing variables, which hinders their applicability in ML. To solve this issue, canonical forms of TNs are parameterized via unitary matrices to simplify the computation of marginals. However, these canonical forms do not apply to circuits, as they can represent factorizations that do not directly map to a known TN. Inspired by the ideas of orthogonality in canonical forms and determinism in circuits enabling tractable maximization, we show how to parameterize squared circuits to overcome their marginalization overhead. Our parameterizations unlock efficient marginalization even in factorizations different from TNs, but encoded as circuits, whose structure would otherwise make marginalization computationally hard. Finally, our experiments on distribution estimation show how our proposed conditions in squared circuits come with no expressiveness loss, while enabling more efficient learning.
format Preprint
id arxiv_https___arxiv_org_abs_2512_17090
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle How to Square Tensor Networks and Circuits Without Squaring Them
Loconte, Lorenzo
Javaloy, Adrián
Vergari, Antonio
Machine Learning
Artificial Intelligence
Squared tensor networks (TNs) and their extension as computational graphs--squared circuits--have been used as expressive distribution estimators, yet supporting closed-form marginalization. However, the squaring operation introduces additional complexity when computing the partition function or marginalizing variables, which hinders their applicability in ML. To solve this issue, canonical forms of TNs are parameterized via unitary matrices to simplify the computation of marginals. However, these canonical forms do not apply to circuits, as they can represent factorizations that do not directly map to a known TN. Inspired by the ideas of orthogonality in canonical forms and determinism in circuits enabling tractable maximization, we show how to parameterize squared circuits to overcome their marginalization overhead. Our parameterizations unlock efficient marginalization even in factorizations different from TNs, but encoded as circuits, whose structure would otherwise make marginalization computationally hard. Finally, our experiments on distribution estimation show how our proposed conditions in squared circuits come with no expressiveness loss, while enabling more efficient learning.
title How to Square Tensor Networks and Circuits Without Squaring Them
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2512.17090