Group-Algebraic Tensors: Provably-optimal Equivariant Learning and Physical Symmetry Discovery

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Hoyos, Paulina, Ubaru, Shashanka, Huh, Dongsung, Kalantzis, Vasileios, Clarkson, Kenneth L., Kilmer, Misha, Avron, Haim, Horesh, Lior
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914581485453312
author Hoyos, Paulina
Ubaru, Shashanka
Huh, Dongsung
Kalantzis, Vasileios
Clarkson, Kenneth L.
Kilmer, Misha
Avron, Haim
Horesh, Lior
author_facet Hoyos, Paulina
Ubaru, Shashanka
Huh, Dongsung
Kalantzis, Vasileios
Clarkson, Kenneth L.
Kilmer, Misha
Avron, Haim
Horesh, Lior
contents We introduce the $\star_G$ tensor algebra, in which any finite group $G$ defines the multiplication rule, making equivariance an intrinsic algebraic property rather than an architectural constraint. The framework rests on three machine-verified theoretical pillars: (i)~an Eckart-Young optimality guarantee for the $\star_G$-SVD: the first such result for symmetry-preserving tensor approximation, exact and polynomial-time; (ii)~a Kronecker factorization that composes multiple symmetries by replacing $F_G$ with $F_{G_1} \otimes F_{G_2}$ with no architectural redesign; and (iii)~a 600-line Lean~4 formalization of the $\star_G$ algebra. The framework provides capabilities that equivariant neural networks (ENNs) structurally cannot: a closed-form per-irreducible-representation decomposition of every prediction, and data-driven discovery of the symmetry group that best fits a dataset. As a non-trivial empirical demonstration, decomposing QM9 molecular geometry over the chiral octahedral subgroup of SO(3) recovers the Wigner--Eckart selection rules of angular momentum from data alone, with no quantum mechanical input: scalar properties are A$_1$-dominated, dipole components are T$_1$-dominated, the isotropic polarizability is uniquely insensitive to $l\!=\!1$ as the rank-2-trace decomposition $l\!=\!0 \oplus l\!=\!2$ requires, and the T$_1$/A$_1$ predictive-power ratio separates vector observables from scalar observables by a factor of five. On full QM9 (130{,}831 molecules), $\star_G$-SVD with ridge regression provides closed form predictions at $\sim50-90\times$ fewer parameters than parameter-matched MLPs. Algebraic equivariance thus complements architectural equivariance not as a faster-better-cheaper alternative but as a different mathematical affordance: provably-optimal symmetry-preserving compression, per-irrep interpretability, and data-driven physical discovery.
format Preprint
id arxiv_https___arxiv_org_abs_2605_20440
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Group-Algebraic Tensors: Provably-optimal Equivariant Learning and Physical Symmetry Discovery
Hoyos, Paulina
Ubaru, Shashanka
Huh, Dongsung
Kalantzis, Vasileios
Clarkson, Kenneth L.
Kilmer, Misha
Avron, Haim
Horesh, Lior
Machine Learning
Artificial Intelligence
Rings and Algebras
We introduce the $\star_G$ tensor algebra, in which any finite group $G$ defines the multiplication rule, making equivariance an intrinsic algebraic property rather than an architectural constraint. The framework rests on three machine-verified theoretical pillars: (i)~an Eckart-Young optimality guarantee for the $\star_G$-SVD: the first such result for symmetry-preserving tensor approximation, exact and polynomial-time; (ii)~a Kronecker factorization that composes multiple symmetries by replacing $F_G$ with $F_{G_1} \otimes F_{G_2}$ with no architectural redesign; and (iii)~a 600-line Lean~4 formalization of the $\star_G$ algebra. The framework provides capabilities that equivariant neural networks (ENNs) structurally cannot: a closed-form per-irreducible-representation decomposition of every prediction, and data-driven discovery of the symmetry group that best fits a dataset. As a non-trivial empirical demonstration, decomposing QM9 molecular geometry over the chiral octahedral subgroup of SO(3) recovers the Wigner--Eckart selection rules of angular momentum from data alone, with no quantum mechanical input: scalar properties are A$_1$-dominated, dipole components are T$_1$-dominated, the isotropic polarizability is uniquely insensitive to $l\!=\!1$ as the rank-2-trace decomposition $l\!=\!0 \oplus l\!=\!2$ requires, and the T$_1$/A$_1$ predictive-power ratio separates vector observables from scalar observables by a factor of five. On full QM9 (130{,}831 molecules), $\star_G$-SVD with ridge regression provides closed form predictions at $\sim50-90\times$ fewer parameters than parameter-matched MLPs. Algebraic equivariance thus complements architectural equivariance not as a faster-better-cheaper alternative but as a different mathematical affordance: provably-optimal symmetry-preserving compression, per-irrep interpretability, and data-driven physical discovery.
title Group-Algebraic Tensors: Provably-optimal Equivariant Learning and Physical Symmetry Discovery
topic Machine Learning
Artificial Intelligence
Rings and Algebras
url https://arxiv.org/abs/2605.20440