Arrow Research search
Back to I&C

I&C 2017

Quantitative simulations by matrices

Journal Article journal-article Computer Science · Theoretical Computer Science

Abstract

We introduce notions of simulation between semiring-weighted automata as models of quantitative systems. Our simulations are instances of the categorical/coalgebraic notions previously studied by Hasuo—hence soundness against language inclusion comes for free—but are concretely presented as matrices that are subject to linear inequality constraints. Pervasiveness of these formalisms allows us to exploit existing algorithms in: searching for a simulation, and hence verifying quantitative correctness that is formulated as language inclusion. Transformations of automata that aid search for simulations are introduced, too. This verification workflow is implemented for the plus-times and max-plus semirings. Furthermore, an extension to weighted tree automata is presented and implemented.

Authors

Keywords

  • Kleisli category
  • Simulation
  • Language inclusion
  • Weighted automaton
  • Tropical semiring

Context

Venue
Information and Computation
Archive span
1987-2026
Indexed papers
3021
Paper id
993168367112794243
v2026.09.13