A Model-checking Approach to Reduce Spiking Neural Networks

Elisabetta de Maria (1) , Daniel Gaffé (2) , Annie Ressouche (3) , Cédric Girard Riboulleau (3)

1 Laboratoire d'Informatique, Signaux, et Systèmes de Sophia-Antipolis (I3S) / Equipe BIOINFO
2 LEAT - Laboratoire d'Electronique, Antennes et Télécommunications
3 STARS - Spatio-Temporal Activity Recognition Systems

In this paper we formalize Boolean Probabilistic Leaky Integrate and Fire Neural Networks as Discrete-Time Markov Chains using the language PRISM. In our models, the probability for neurons to emit spikes is driven by the difference between their membrane potential and their firing threshold. The potential value of each neuron is computed taking into account both the current input signals and the past potential value. Taking advantage of this modeling, we propose a novel algorithm which aims at reducing the number of neurons and synaptical connections of a given network. The reduction preserves the desired dynamical behavior of the network, which is formalized by means of temporal logic formulas and verified thanks to the PRISM model checker.

Mots clés: Neural Spiking Networks; Probabilistic Models; Temporal Logic; Model Checking; Network Reduction

Domaines: Bio-informatique [q-bio.QM]

Elisabetta de Maria, Daniel Gaffé, Annie Ressouche, Cédric Girard Riboulleau. A Model-checking Approach to Reduce Spiking Neural Networks. BIOINFORMATICS 2018 - 9th International Conference on Bioinformatics Models, Methods and Algorithms, Jan 2018, Funchal Madeira, Portugal. pp.1-8, ⟨10.5220/0006572000890096⟩. ⟨hal-01638248⟩