Conditional Clifford-Steerable CNNs with Complete Kernel Basis for PDE Modeling

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Szarvas, Bálint László, Zhdanov, Maksim
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914096062922752
author Szarvas, Bálint László
Zhdanov, Maksim
author_facet Szarvas, Bálint László
Zhdanov, Maksim
contents Clifford-Steerable CNNs (CSCNNs) provide a unified framework that allows incorporating equivariance to arbitrary pseudo-Euclidean groups, including isometries of Euclidean space and Minkowski spacetime. In this work, we demonstrate that the kernel basis of CSCNNs is not complete, thus limiting the model expressivity. To address this issue, we propose Conditional Clifford-Steerable Kernels, which augment the kernels with equivariant representations computed from the input feature field. We derive the equivariance constraint for these input-dependent kernels and show how it can be solved efficiently via implicit parameterization. We empirically demonstrate an improved expressivity of the resulting framework on multiple PDE forecasting tasks, including fluid dynamics and relativistic electrodynamics, where our method consistently outperforms baseline methods.
format Preprint
id arxiv_https___arxiv_org_abs_2510_14007
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Conditional Clifford-Steerable CNNs with Complete Kernel Basis for PDE Modeling
Szarvas, Bálint László
Zhdanov, Maksim
Machine Learning
Artificial Intelligence
Clifford-Steerable CNNs (CSCNNs) provide a unified framework that allows incorporating equivariance to arbitrary pseudo-Euclidean groups, including isometries of Euclidean space and Minkowski spacetime. In this work, we demonstrate that the kernel basis of CSCNNs is not complete, thus limiting the model expressivity. To address this issue, we propose Conditional Clifford-Steerable Kernels, which augment the kernels with equivariant representations computed from the input feature field. We derive the equivariance constraint for these input-dependent kernels and show how it can be solved efficiently via implicit parameterization. We empirically demonstrate an improved expressivity of the resulting framework on multiple PDE forecasting tasks, including fluid dynamics and relativistic electrodynamics, where our method consistently outperforms baseline methods.
title Conditional Clifford-Steerable CNNs with Complete Kernel Basis for PDE Modeling
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2510.14007