Linear cellular automata, duality and sofic groups
Laurent Bartholdi

TL;DR
This paper constructs non-pre-injective, surjective linear cellular automata on non-amenable groups, and establishes duality relations linking injectivity and surjectivity properties in cellular automata over sofic groups.
Contribution
It provides a positive solution to an open problem for non-amenable groups and introduces a duality framework connecting injectivity and surjectivity in linear cellular automata.
Findings
Constructed non-pre-injective, surjective linear cellular automata on non-amenable groups.
Reproved that cellular automata over sofic groups are injective iff post-surjective.
Established duality between kernel and image of matrices over group rings.
Abstract
We produce for arbitrary non-amenable group and field a non-pre-injective, surjective linear cellular automaton. This answers positively Open Problem (OP-14) in Ceccherini-Silberstein and Coornaert's monograph "Cellular Automata and Groups". We also reprove in a direct manner, for linear cellular automata, the result by Capobianco, Kari and Taati that cellular automata over sofic groups are injective if and only if they are post-surjective. These results come from considerations related to matrices over group rings: we prove that a matrix's kernel and the image of its adjoint are mutual orthogonals of each other. This gives rise to a notion of "dual cellular automaton", which is pre-injective if and only if the original cellular automaton is surjective, and is injective if and only if the original cellular automaton is post-surjective.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsCellular Automata and Applications · Computability, Logic, AI Algorithms · Mathematical Dynamics and Fractals
