Accelerating Newton-Schulz Iteration for Orthogonalization via Chebyshev-type Polynomials

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Grishina, Ekaterina, Smirnov, Matvey, Rakhuba, Maxim
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910030674001920
author Grishina, Ekaterina
Smirnov, Matvey
Rakhuba, Maxim
author_facet Grishina, Ekaterina
Smirnov, Matvey
Rakhuba, Maxim
contents The problem of computing optimal orthogonal approximation to a given matrix has attracted growing interest in machine learning. Notable applications include the recent Muon optimizer or Riemannian optimization on the Stiefel manifold. Among existing approaches, the Newton-Schulz iteration has emerged as a particularly effective solution, as it relies solely on matrix multiplications and thus achieves high computational efficiency on GPU hardware. Despite its efficiency, the method has inherent limitations - its coefficients are fixed and thus not optimized for a given matrix. In this paper we address this issue by proposing a Chebyshev-optimized version of Newton-Schulz (CANS). Based on the Chebyshev's alternance theorem, we theoretically derive optimal coefficients for the 3-rd order Newton-Schulz iteration and apply a Remez algorithm to compute optimal higher-degree polynomials. We leverage these polynomials to construct controlled approximate orthogonalization schemes, which is of interest in deep learning applications. Practically, we demonstrate the method's effectiveness in two key applications: orthogonalization in the Muon optimizer, and providing an efficient retraction alternative for Riemannian optimization on the Stiefel manifold.
format Preprint
id arxiv_https___arxiv_org_abs_2506_10935
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Accelerating Newton-Schulz Iteration for Orthogonalization via Chebyshev-type Polynomials
Grishina, Ekaterina
Smirnov, Matvey
Rakhuba, Maxim
Numerical Analysis
65F25, 65F60, 53Z50, 68W25
The problem of computing optimal orthogonal approximation to a given matrix has attracted growing interest in machine learning. Notable applications include the recent Muon optimizer or Riemannian optimization on the Stiefel manifold. Among existing approaches, the Newton-Schulz iteration has emerged as a particularly effective solution, as it relies solely on matrix multiplications and thus achieves high computational efficiency on GPU hardware. Despite its efficiency, the method has inherent limitations - its coefficients are fixed and thus not optimized for a given matrix. In this paper we address this issue by proposing a Chebyshev-optimized version of Newton-Schulz (CANS). Based on the Chebyshev's alternance theorem, we theoretically derive optimal coefficients for the 3-rd order Newton-Schulz iteration and apply a Remez algorithm to compute optimal higher-degree polynomials. We leverage these polynomials to construct controlled approximate orthogonalization schemes, which is of interest in deep learning applications. Practically, we demonstrate the method's effectiveness in two key applications: orthogonalization in the Muon optimizer, and providing an efficient retraction alternative for Riemannian optimization on the Stiefel manifold.
title Accelerating Newton-Schulz Iteration for Orthogonalization via Chebyshev-type Polynomials
topic Numerical Analysis
65F25, 65F60, 53Z50, 68W25
url https://arxiv.org/abs/2506.10935