Spiking Neural Networks for Sustainable and High-Performance Computing

Authors:

Abstract
Neuromorphic computing is defined as a non-von Neumann computation, unlike Von Neuman computation which consists of a separate CPU and memory unit, where data and instructions are stored in a unified memory unit. Neuromorphic computing whose structure is inspired by brains composed of neurons and synapses uses spikes as input, capturing the timing and interval between spikes to encode numerical information. it is ideal for systems that require high parallel operation, neuromorphic systems are highly scalable, and event-driven which allows for efficient computation. These neurological and biological mechanisms are modeled in neuromorphic computing systems through spiking neural networks (SNNs), In SNNs the information is coded with time-dependent spikes, which remarkably reduces the power consumption, our proposed solutions result indicate improved learning capacity and reduced power consumption compared to traditional computing approaches.