Neuromorphic Computing with Large Scale Spiking Neural Networks

Abstract

Spiking Neural Networks (SNNs) have emerged as a promising paradigm for biologically inspired computing, offering advantages in energy efficiency, temporal processing, and event-driven computation. As research advances, scaling SNNs to large networks remains a critical challenge, requiring innovations in efficient training algorithms, neuromorphic hardware, and real-world deployment. This survey provides a comprehensive overview of large-scale SNNs, discussing state-of-the-art neuron models, training methodologies, and hardware implementations. We explore key applications in neuroscience, robotics, computer vision, and edge AI, highlighting the advantages and limitations of SNN-based systems. Additionally, we identify open challenges in scalability, energy efficiency, and learning mechanisms, outlining future research directions to bridge the gap between theory and practice. By addressing these challenges, large-scale SNNs have the potential to revolutionize artificial intelligence by providing more efficient, brain-inspired computation frameworks.