Work

Application and Simulation of Neuromorphic Devices for use in Neural Networks

Year: 2018
Type: article
Abstract: Software and hardware artificial neural networks have been increasingly used to solve difficult real world problems.Neural networks have shown successful results in both the High Performance Computing and Big Data fields, where computation is often distributed among clustered hardware, such as CPUs and general purpose Graphic Processing Units (GPUs), to form much larger computing systems.The growing field of neuromorphic computing combines aspects of complex biological systems and classical machine learning methods, accelerated by alternative physical platforms.The field exemplifies a complementary class of neural network computing platforms that aim to emulate neural and synaptic dynamics performed by the human brain with solid-state devices designed to mimic the electro-chemical Chapter 1 Neuromorphic Computing 1.1 An Emerging Discipline The advent of neuromorphic computing has inspired a variety of broad perspectives in memory and computer architectures for use in machine learning applications.Most modern computing platforms are based on the Von Neumann architecture which separates memory and computation into separate hierarchical units (such as caches, registers, control unit, etc.).The decreasing cost and increase in density of transistors, has made the clock-driven Von Neumann architecture an increasingly powerful option in computing.Due to Moores law slowly coming to an end with present transistor and facing physical and quantum limits due to channel size tradeoffs, processing speeds and computing power of modern platforms are dampening.Distributed platforms, while they offer the possibility of infinite horizontal scalability, come at increasing cost of management and speed with Moores law.This is
Author: Wenke, Sam
Language: English
Open Access status: green
OpenAlex ID: W7054999867
FWCI:
Cites: 23
Cited by:
Related to: 9