Riemannian Optimization for Hadamard Products of Low-Rank Matrices

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jawanpuria, Pratik, Chandresh, Ankish, Mishra, Bamdev
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918534318129152
author Jawanpuria, Pratik
Chandresh, Ankish
Mishra, Bamdev
author_facet Jawanpuria, Pratik
Chandresh, Ankish
Mishra, Bamdev
contents The elementwise Hadamard product of two low-rank matrices provides a parameter-efficient model for data with multiplicative structure, but its modeling is challenging due to the presence of additional symmetries under coupled row/column scalings between the two factors. In order to leverage the geometry of the space, we formulate the learning of such matrices as optimization on a Riemannian quotient manifold. We propose a novel block-diagonal Riemannian metric derived from the pullback of the Frobenius inner product. The metric is shown to be invariant under the full symmetry group. We develop a Riemannian gradient descent algorithm that uses a tuning-free Gauss--Newton step size and scales linearly in the number of observed entries per iteration. Experiments on real and synthetic datasets illustrate the efficacy of our proposed Riemannian approach.
format Preprint
id arxiv_https___arxiv_org_abs_2606_01216
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Riemannian Optimization for Hadamard Products of Low-Rank Matrices
Jawanpuria, Pratik
Chandresh, Ankish
Mishra, Bamdev
Machine Learning
Optimization and Control
The elementwise Hadamard product of two low-rank matrices provides a parameter-efficient model for data with multiplicative structure, but its modeling is challenging due to the presence of additional symmetries under coupled row/column scalings between the two factors. In order to leverage the geometry of the space, we formulate the learning of such matrices as optimization on a Riemannian quotient manifold. We propose a novel block-diagonal Riemannian metric derived from the pullback of the Frobenius inner product. The metric is shown to be invariant under the full symmetry group. We develop a Riemannian gradient descent algorithm that uses a tuning-free Gauss--Newton step size and scales linearly in the number of observed entries per iteration. Experiments on real and synthetic datasets illustrate the efficacy of our proposed Riemannian approach.
title Riemannian Optimization for Hadamard Products of Low-Rank Matrices
topic Machine Learning
Optimization and Control
url https://arxiv.org/abs/2606.01216