Examining Kempe equivalence via commutative algebra
Hidefumi Ohsugi, Akiyoshi Tsuchiya
TL;DR
This work builds an algebraic framework to study Kempe equivalence of graph colorings by mapping colorings to binomial ideals arising from stable sets and 2-colorings. It proves that Kempe equivalence can be detected via membership in the 2-coloring ideal $J_G$ (equivalently in the quadratic part of the stable-set toric ideal), and extends this with the Kempe ideal $K_G$ to obtain complete representatives for $k$-Kempe classes using Gröbner-basis techniques. A key result is that the Hilbert function of the quotient by $K_G$ encodes the number of Kempe classes across induced subgraphs, enabling computation via algebraic invariants. The paper also provides a suite of algebraic algorithms for deciding Kempe equivalence, enumerating Kempe classes, and constructing explicit sequences of Kempe switches, thereby offering a uniform, computable framework for Kempe theory grounded in commutative algebra and toric geometry.
Abstract
Kempe equivalence is a classical and important notion on vertex coloring in graph theory. In the present paper, we introduce several ideals associated with graphs and provide a method to determine whether two $k$-colorings are Kempe equivalent via commutative algebra. Moreover, we give a way to compute all $k$-colorings of a graph up to Kempe equivalence by virtue of the algebraic technique on Gröbner bases. As a consequence, the number of $k$-Kempe classes can be computed by using Hilbert functions. Finally, we introduce several algebraic algorithms related to Kempe equivalence.
