In this paper, we define spiking neural P circuits (SN P circuits) as an acyclic variant of spiking neural P systems. We then study how well genetic algorithms (GA) are able to find an SN P circuit that computes a given Boolean function, possibly partially defined. The proposed technique can be used to find SN P circuits that solve binary classification problems. We performed several computer experiments, testing different mutation operators and several combinations of hyperparameter values. The preliminary results obtained show that the probability of success of GA strongly depends upon the structure (in particular, the algebraic degree and the number of input/output variables) of the Boolean function to be computed.

Leporati, A., Rovida, L. (2025). An evolutionary approach to the design of spiking neural P circuits. JOURNAL OF MEMBRANE COMPUTING [10.1007/s41965-024-00180-x].

An evolutionary approach to the design of spiking neural P circuits

Leporati A.
;
Rovida L.
2025

Abstract

In this paper, we define spiking neural P circuits (SN P circuits) as an acyclic variant of spiking neural P systems. We then study how well genetic algorithms (GA) are able to find an SN P circuit that computes a given Boolean function, possibly partially defined. The proposed technique can be used to find SN P circuits that solve binary classification problems. We performed several computer experiments, testing different mutation operators and several combinations of hyperparameter values. The preliminary results obtained show that the probability of success of GA strongly depends upon the structure (in particular, the algebraic degree and the number of input/output variables) of the Boolean function to be computed.
Articolo in rivista - Articolo scientifico
Binary classification; Boolean functions; Evolutionary algorithms; Spiking neural P systems;
English
14-feb-2025
2025
119916
none
Leporati, A., Rovida, L. (2025). An evolutionary approach to the design of spiking neural P circuits. JOURNAL OF MEMBRANE COMPUTING [10.1007/s41965-024-00180-x].
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/548202
Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? 0
Social impact