Study of Spiking Neural Network Architecture for Neuromorphic Computing

Abstract

Deep learning's progress has resulted in a multi-layered character in a variety of applications in this field. Artificial Neural Networks are becoming the old procedure in the wide region of computing, using fifty-year-old principal concepts and models. Scientists are now presenting 3rd generation intelligent models in this current computer era. The brain like a big computer, routes information that it receives from the senses and body, and sends communications back to the body. But the brain can do much more than a machine can: humans think and experience reactions with their brain, and are the root of human intelligence. The 3rd generation Spiking Neural Network bridges the gap between deep learning, machine learning, and neuroscience in a biological approach, allowing neuroscience and machine learning to work together to achieve high-level computing efficiency using the neuromorphic computing. Spiking Neural Networks promise to make use of spikes, which are discrete functions that occur at predictable emphases rather than continuous values so that they are hardware implementable as well.