The Price of Freedom: Exploring Expressivity and Runtime Tradeoffs in Equivariant Tensor Products

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Xie, YuQing, Daigavane, Ameya, Kotak, Mit, Smidt, Tess
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909689802915840
author Xie, YuQing
Daigavane, Ameya
Kotak, Mit
Smidt, Tess
author_facet Xie, YuQing
Daigavane, Ameya
Kotak, Mit
Smidt, Tess
contents $E(3)$-equivariant neural networks have demonstrated success across a wide range of 3D modelling tasks. A fundamental operation in these networks is the tensor product, which interacts two geometric features in an equivariant manner to create new features. Due to the high computational complexity of the tensor product, significant effort has been invested to optimize the runtime of this operation. For example, Luo et al. (2024) recently proposed the Gaunt tensor product (GTP) which promises a significant speedup. In this work, we provide a careful, systematic analysis of a number of tensor product operations. In particular, we emphasize that different tensor products are not performing the same operation. The reported speedups typically come at the cost of expressivity. We introduce measures of expressivity and interactability to characterize these differences. In addition, we realized the original implementation of GTP can be greatly simplified by directly using a spherical grid at no cost in asymptotic runtime. This spherical grid approach is faster on our benchmarks and in actual training of the MACE interatomic potential by 30%. Finally, we provide the first systematic microbenchmarks of the various tensor product operations. We find that the theoretical runtime guarantees can differ wildly from empirical performance, demonstrating the need for careful application-specific benchmarking. Code is available at https://github.com/atomicarchitects/PriceofFreedom.
format Preprint
id arxiv_https___arxiv_org_abs_2506_13523
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle The Price of Freedom: Exploring Expressivity and Runtime Tradeoffs in Equivariant Tensor Products
Xie, YuQing
Daigavane, Ameya
Kotak, Mit
Smidt, Tess
Machine Learning
Artificial Intelligence
$E(3)$-equivariant neural networks have demonstrated success across a wide range of 3D modelling tasks. A fundamental operation in these networks is the tensor product, which interacts two geometric features in an equivariant manner to create new features. Due to the high computational complexity of the tensor product, significant effort has been invested to optimize the runtime of this operation. For example, Luo et al. (2024) recently proposed the Gaunt tensor product (GTP) which promises a significant speedup. In this work, we provide a careful, systematic analysis of a number of tensor product operations. In particular, we emphasize that different tensor products are not performing the same operation. The reported speedups typically come at the cost of expressivity. We introduce measures of expressivity and interactability to characterize these differences. In addition, we realized the original implementation of GTP can be greatly simplified by directly using a spherical grid at no cost in asymptotic runtime. This spherical grid approach is faster on our benchmarks and in actual training of the MACE interatomic potential by 30%. Finally, we provide the first systematic microbenchmarks of the various tensor product operations. We find that the theoretical runtime guarantees can differ wildly from empirical performance, demonstrating the need for careful application-specific benchmarking. Code is available at https://github.com/atomicarchitects/PriceofFreedom.
title The Price of Freedom: Exploring Expressivity and Runtime Tradeoffs in Equivariant Tensor Products
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2506.13523