Spontaneous Dynamics of Asymmetric Random Recurrent Spiking Neural Networks

By Samanwoy Ghosh-Dastidar, Hojjat Adeli. Neural Computation, 2004

748 citations recorded by the source corpus.

In this letter, we study the effect of a unique initial stimulation on random recurrent networks of leaky integrate-and-fire neurons. Indeed, given a stochastic connectivity, this so-called spontaneous mode exhibits various nontrivial dynamics. This study is based on a mathematical formalism that allows us to examine the variability of the afterward dynamics according to the parameters of the weight distribution. Under the independence hypothesis (e.g., in the case of very large networks), we are able to compute the average number of neurons that fire at a given timethe spiking activity. In accordance with numerical simulations, we prove that this spiking activity reaches a steady state. We characterize this steady state and explore the transients.

Subfields: Cognitive Neuroscience, Electrical and Electronic Engineering, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/659dd503b7cbaf297395bdecc6764fba8ef9deaf

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Electrical and Electronic Engineering

In Statistical and Nonlinear Physics

Citations in the corpus

Why Neurons Have Thousands of Synapses, a Theory of Sequence Memory in Neocortex (2015, background): “A number of papers have studied spiking neuron models (Ghosh-Dastidar and Adeli, 2009; Maass, 1997) in the context of sequences.”
Biologically inspired sleep algorithm for increased generalization and adversarial robustness in deep neural networks (2020, background): “In short, SNNs are composed of spiking neurons and model the information transformation and the dependence on exact timing of spikes that occurs in biological networks (Ghosh-Dastidar & Adeli (2009)).”
Neuromorphic computing using non-volatile memory (2017, background): “In an SNN, spiking of the pre-synaptic neuron (electrical current) modifies the membrane potential (electrical voltage) of the post-synaptic neuron, by an amount determined by the ‘synaptic weight’ (electrical conductance), leading eventually to a post-synaptic spike through a leaky-integrate-and-fi”
Deep Learning in Spiking Neural Networks (2018, background): “The introduction of SNNs in the last few decades, as a powerful third generation neural network [50], has encouraged many studies with the focus on biologically motivated approaches for pattern recognition [62], [63].”
Minerva: Enabling Low-Power, Highly-Accurate Deep Neural Network Accelerators (2016, background): “Spiking Neural Networks (SNNs) [57, 58, 59, 60] are popular in ASIC designs, largely due to their compact mapping onto fully-parallel hardware.”