English

Computation of lattice isomorphisms and the integral matrix similarity problem

Number Theory 2022-09-01 v3 Group Theory Rings and Algebras

Abstract

Let KK be a number field, let AA be a finite-dimensional KK-algebra, let J(A)\mathrm{J}(A) denote the Jacobson radical of AA, and let Λ\Lambda be an OK\mathcal{O}_{K}-order in AA. Suppose that each simple component of the semisimple KK-algebra A/J(A)A/{\mathrm{J}(A)} is isomorphic to a matrix ring over a field. Under this hypothesis on AA, we give an algorithm that given two Λ\Lambda-lattices XX and YY, determines whether XX and YY are isomorphic, and if so, computes an explicit isomorphism XYX \rightarrow Y. This algorithm reduces the problem to standard problems in computational algebra and algorithmic algebraic number theory in polynomial time. As an application, we give an algorithm for the following long-standing problem: given a number field KK, a positive integer nn and two matrices A,BMatn(OK)A,B \in \mathrm{Mat}_{n}(\mathcal{O}_{K}), determine whether AA and BB are similar over OK\mathcal{O}_{K}, and if so, return a matrix CGLn(OK)C \in \mathrm{GL}_{n}(\mathcal{O}_{K}) such that B=CAC1B= CAC^{-1}. We give explicit examples that show that the implementation of the latter algorithm for OK=Z\mathcal{O}_{K}=\mathbb{Z} vastly outperforms implementations of all previous algorithms, as predicted by our complexity analysis.

Keywords

Cite

@article{arxiv.2202.03526,
  title  = {Computation of lattice isomorphisms and the integral matrix similarity problem},
  author = {Werner Bley and Tommy Hofmann and Henri Johnston},
  journal= {arXiv preprint arXiv:2202.03526},
  year   = {2022}
}

Comments

37 pages; v3 revised and accepted version to appear in Forum of Mathematics, Sigma; v2 has many minor corrections and changes