Author_Institution :
Dept. of Electr. Eng., Tufts Univ., Medford, MA, USA
Abstract :
Neural networks is a very popular emerging technology that has applications in a wide variety of fields such as: electrical and computer engineering, signal processing, computer science, cognitive science, neurobiology, optics, mathematics, and physics. Properties of this multidisciplinary field, originally inspired by biological systems, are adaptiveness and self organization, massively parallel nature, robustness and fault- and noise-tolerance. This paper, after presenting an overview of neural networks, their properties and advantages over conventional computers, will discuss how novel neural network architectures and implementations can be included in larger systems to provide solutions to scientific and engineering problems. An example of a neural network hearing model is discussed where neural networks and time-frequency analysis are used to extract essential information from complex inputs
Keywords :
engineering computing; neural nets; real-time systems; adaptiveness; neural network architectures; neural network hearing model; neural networks; real-time engineering systems; self organization; time-frequency analysis; Application software; Biology computing; Computer networks; Electrical engineering computing; Neural networks; Optical computing; Optical signal processing; Physics computing; Real time systems; Systems engineering and theory;
