English

A Weighted Quiver Kernel using Functor Homology

Machine Learning 2020-09-29 v1 Machine Learning

Abstract

In this paper, we propose a new homological method to study weighted directed networks. Our model of such networks is a directed graph QQ equipped with a weight function ww on the set Q1Q_{1} of arrows in QQ. We require that the range WW of our weight function is equipped with an addition or a multiplication, i.e., WW is a monoid in the mathematical terminology. When WW is equipped with a representation on a vector space MM, the standard method of homological algebra allows us to define the homology groups H(Q,w;M)H_{*}(Q,w;M). It is known that when QQ has no oriented cycles, Hn(Q,w;M)=0H_{n}(Q,w;M)=0 for n2n\ge 2 and H1(Q,w;M)H_{1}(Q,w;M) can be easily computed. This fact allows us to define a new graph kernel for weighted directed graphs. We made two sample computations with real data and found that our method is practically applicable.

Keywords

Cite

@article{arxiv.2009.12928,
  title  = {A Weighted Quiver Kernel using Functor Homology},
  author = {Manohar Kaul and Dai Tamaki},
  journal= {arXiv preprint arXiv:2009.12928},
  year   = {2020}
}

Comments

23 pages, 5 figures

R2 v1 2026-06-23T18:49:43.574Z