Algorithm for computing $\mu$-bases of univariate polynomials

TL;DR

This paper introduces a novel algorithm for computing $$-bases of univariate polynomial syzygy modules, offering a linear algebra-based approach with proven correctness and complexity analysis, and compares its performance with existing methods.

Contribution

It presents a new, conceptually different algorithm for $$-basis computation over any field, with a self-contained proof of correctness and complexity analysis.

Findings

The new algorithm has a worst-case complexity of $O(d^2n+d^3+n^2)$.

Experimental results show the new algorithm outperforms existing methods for large $n$ or $d$.

The algorithm provides an alternative proof of the existence and freeness of the $$-basis.

Abstract

We present a new algorithm for computing a $\mu$-basis of the syzygy module of $n$ polynomials in one variable over an arbitrary field $\mathbb{K}$. The algorithm is conceptually different from the previously-developed algorithms by Cox, Sederberg, Chen, Zheng, and Wang for $n=3$, and by Song and Goldman for an arbitrary $n$. It involves computing a "partial" reduced row-echelon form of a $ (2d+1)\times n(d+1)$ matrix over $\mathbb{K}$, where $d$ is the maximum degree of the input polynomials. The proof of the algorithm is based on standard linear algebra and is completely self-contained. It includes a proof of the existence of the $\mu$-basis and as a consequence provides an alternative proof of the freeness of the syzygy module. The theoretical (worst case asymptotic) computational complexity of the algorithm is $O(d^2n+d^3+n^2)$. We have implemented this algorithm (HHK) and the one developed by Song and Goldman (SG). Experiments on random inputs indicate that SG gets faster than HHK when $d$ gets sufficiently large for a fixed $n$, and that HHK gets faster than SG when $n$ gets sufficiently large for a fixed $d$.