A statistical approach to the identification of diploid cellular automata based on incomplete observations. (December 2019)

• Record Type: Journal Article
• Authors: Bołt, Witold; Bołt, Aleksander; Wolnik, Barbara; Baetens, Jan M.; De Baets, Bernard
• Is Part Of: Bio systems. Volume 186(2019)
• Publication Date: 2019-12
• Subjects: Stochastic cellular automata -- Diploid cellular automata -- Parameter estimation -- System identification; Biological systems -- Periodicals; Biology -- Periodicals; Biology -- Periodicals; Evolution -- Periodicals; Biologie -- Périodiques; Évolution -- Périodiques; 570
• DOI: 10.1016/j.biosystems.2019.103976
• Languages: English
• ISSNs: 0303-2647

Abstract

Abstract: In this paper, the identification problem of diploid cellular automata is considered, in which, based on a series of incomplete observations, the underlying cellular automaton rules and the states of missing cell states are to be uncovered. An algorithm for identifying the rule, based on a statistical parameter estimation method using a normal distribution approximation, is presented. In addition, an algorithm for filling the missing cell states is formulated. The accuracy of these methods is examined in a series of computational experiments.