Resumen

DIAZ, Nestor  y  TISCHER, Irene. Mining stochastic cellular automata to solve density classification task in two dimensions. Dyna rev.fac.nac.minas [online]. 2020, vol.87, n.215, pp.39-46.  Epub 04-Ene-2021. ISSN 0012-7353.  https://doi.org/10.15446/dyna.v87n215.83200.

Density Classification Task (DCT) is a well-known problem, where the main goal is to build a cellular automaton whose local rule gives rise to emergent global coordination. We describe the methods used to identify new cellular automata that solve this problem. Our approach identifies both the neighborhood and its stochastic rule using a dataset of initial configurations that covers in a predefined way the full range of densities in DCT. We compare our results with some models currently available in the field. In some cases, our models show better performance than the best solution reported in the literature, with efficacy of 0.842 for datasets with uniform distribution around the critical density. Tests were carried out in datasets of diverse lattice sizes and sampling conditions. Finally, by a statistical non-parametric test, we demonstrate that there are no significant differences between our identified cellular automata and the best-known model.

Palabras clave : automated model design; computational framework; machine learning; genetic algorithm; Friedman test; Nemenyi’s post-hoc test.

        · resumen en Español     · texto en Inglés     · Inglés ( pdf )