Neuromorphic and Brain-Inspired VLSI Architectures

Abstract

ABSTRACT Neuromorphic computing and brain-inspired VLSI architectures have emerged as promising approaches to overcome the limitations of conventional computing systems. By mimicking the structure and functionality of the human brain, these architectures aim to achieve low-power, high-speed, and fault tolerant computation. This review paper presents the recent developments in neuromorphic VLSI systems, including neuron and synapse circuit designs, spiking neural networks (SNNs), and memory-centric architectures. Various design challenges, such as scalability, device variability, and energy efficiency, are discussed. Furthermore, the paper highlights potential applications of neuromorphic systems in artificial intelligence, robotics, and edge computing. The survey also includes comparisons of different brain-inspired architectures and identifies key research directions for future work.

KEYWORDS: Neuromorphic Computing, Brain-Inspired VLSI, Spiking Neural Networks, Low-Power Circuits, Synaptic Devices, Memristors