Paintbucket on graphs is PSPACE-complete
Ethan J. Saunders, Peter Selinger
TL;DR
This work proves that Paintbucket on graphs, a natural generalization where players flip an opponent's connected color class on a simple graph, is PSPACE-complete. The authors reduce from avoider-enforcer games by constructing a polynomial-size bipartite graph \(G_{K}{(C, {\cal A})}\) with a parameter \(K \ge |C|+2\) that faithfully simulates any avoider-enforcer position \((C, {\cal A})\) via a parity-dependent argument. They introduce an intended play framework and show that any deviation (‘shenanigan’) leads to an immediate loss, ensuring the reduction's correctness through a robust simulation. They prove a key proposition linking the winner in \(G_{K}{(C, {\cal A})}\) to the winner in \((C, {\cal A})\) depending on the parity of \(|C|\), establishing PSPACE-hardness, whilePaintbucket on graphs remains in PSPACE. The paper concludes with an open question regarding the PSPACE status of the original grid-based Paintbucket version.
Abstract
The game of Paintbucket was recently introduced by Amundsen and Erickson. It is played on a rectangular grid of black and white pixels. The players alternately fill in one of their opponent's connected components with their own color, until the entire board is just a single color. The player who makes the last move wins. It is not currently known whether there is a simple winning strategy for Paintbucket. In this paper, we consider a natural generalization of Paintbucket that is played on an arbitrary simple graph, and we show that the problem of determining the winner in a given position of this generalized game is PSPACE-complete.
