AG-2024.04-154·cs.DM·cross-listed: cs.DCgr-qcmath.DSnlin.CG
Space-time deterministic graph rewriting
Authors
- Pablo Arrighi
- Marin Costes
- Gilles Dowek
- Luidnel Maignan
Abstract
We study non-terminating graph rewriting models, whose local rules are applied non-deterministically -- and yet enjoy a strong form of determinism, namely space-time determinism. Of course in the case of terminating computation it is well-known that the mess introduced by asynchronous rule applications may not matter to the end result, as confluence conspires to produce a unique normal form. In the context of non-terminating computation however, confluence is a very weak property, and (almost) synchronous rule applications is always preferred e.g. when it comes to simulating dynamical systems. Here we provide sufficient conditions so that asynchronous local rule applications conspire to produce well-determined events in the space-time unfolding of the graph, regardless of their application orders. Our first example is an asynchronous simulation of a dynamical system. Our second example features time dilation, in the spirit of general relativity.
Submitted
8 April 20242 years ago
Version
v1
License
CC-BY-4.0
DOI
10.48550/arXiv.2404.05838
Summary
When graph-rewriting rules are applied in any order, certain systems still produce deterministic patterns in space-time—like a chaotic kitchen where different cooking orders still yield the same meal.
- The paper identifies conditions under which asynchronous (uncoordinated) rule applications create deterministic outcomes in non-terminating systems, solving a problem that matters for simulating physical dynamics.
- Unlike traditional confluence in computer science, which guarantees one final answer, this 'space-time determinism' guarantees that specific events happen at predictable locations and times regardless of scheduling—enabling simulation of relativistic systems.
- Two concrete examples demonstrate the idea: an asynchronous dynamical system simulator and a model exhibiting time dilation effects, suggesting the framework could unify computation with physics simulations.
curious · generated by claude-haiku-4-5
Chat with this PDF
Ask questions, probe assumptions, request a plain-English summary. Answers cite sections from the preprint itself.
Community
Questions and answers about this paper from other readers. No formal peer review — just a place to think out loud.