A deep learning based surrogate model for the parameter identification problem in probabilistic cellular automaton epidemic models.

Journal: Computer methods and programs in biomedicine
Published Date:

Abstract

BACKGROUND AND OBJECTIVE: an accurate estimation of the epidemiological model coefficients helps understand the basic principles of disease spreading. Some studies showed that dozens of hours are needed to simulate the traditional probabilistic cellular automaton (PCA) model, and dozens of hours are spent for a fine-tuning of the system. Here, we propose a deep learning-based surrogate model to mimic a PCA model to reduce the simulations' computational time, maintaining an equivalent precision in the estimates.

Authors

  • F H Pereira
    Universidade Nove de Julho, Informatics and Knowledge Management Graduate Program, PPGI-UNINOVE, São Paulo, SP, Brazil; Universidade Nove de Julho, Industrial Engineering Graduate Program, PPGEP-UNINOVE, São Paulo, SP, Brazil. Electronic address: fabiohp@uni9.pro.br.
  • P H T Schimit
    Universidade Nove de Julho, Informatics and Knowledge Management Graduate Program, PPGI-UNINOVE, São Paulo, SP, Brazil.
  • F E Bezerra
    Universidade Nove de Julho, Industrial Engineering Graduate Program, PPGEP-UNINOVE, São Paulo, SP, Brazil.