Toward Cognitive Machines: Evaluating Single Device Based Spiking Neural Networks for Brain-Inspired Computing
Faisal Bashir, Ali Alzahrani, Haider Abbas, Furqan Zahoor
Abstract
A brain-inspired computing paradigm known as “neuromorphic computing” seeks to replicate the information processing processes of biological neural systems in order to create computing systems that are effective, low-power, and adaptable. Spiking neural networks (SNNs) based on a single device are at the forefront of brain-inspired computing, which aims to mimic the processing powers of the human brain. Neuromorphic devices, which enable the hardware implementation of artificial neural networks (ANNs), are at the heart of neuromorphic computing. These devices replicate the dynamics and functions of neurons and synapses. This mini-review assesses the latest advancements in neuromorphic computing, with an emphasis on small, energy-efficient devices that mimic biological synapses and neurons. Key neuromorphic functions like spike-timing-dependent plasticity, multistate storage, and dynamic filtering are demonstrated by a variety of single-device models, such as memristors, transistors, and magnetic and ferroelectric devices. The integrate-and-fire (IF) neuron is a key model in these systems because it allows for mathematical analysis while successfully capturing key aspects of neural processing. This review examines the potential of SNNs for scalable, low-power neuromorphic computing applications, highlighting both the benefits and constraints of implementing them with single-device architectures. This review highlights the increasing importance of single-device SNNs in the creation of effective, flexible cognitive devices.