Fuzzy Spiking Neural Networks

Spiking neurons represent the latest generation of neural networks designed to be more biologically realistic than more traditional artificial neural networks. This book provides a hands-on historical perspective on the development of neural networks, and investigates state-of-the-art methods for designing spiking neural networks to solve complex engineering problems. In particular, the book discusses the trade-off between biological plausibility and computational complexity of the numerous alternative models for the electrical properties of the neuron membrane, the electro-chemical interactions at the synapse, and for implementing learning. Often there is a trade-off between the sizes of the networks that can be implemented, and the complexity of the individual neurons in the network. The book investigates the definite lack of a rationale for organising all the different models in a network sense. This investigation has culminated in the development of the fuzzy spiking neural network (FSNN), which utilises fuzzy logic as a rationale for the deployment of biological models within a spiking neural network.

Autorentext
Cornelius Glackin received his B.Sc. degree in Mathematics from Coventry University in 1994. At the University of Ulster he obtained his M.Sc. in Computing and Intelligent Systems in 2004, and his PhD in 2009. He is currently working as a Research Associate within the Intelligent Systems Research Centre at the University of Ulster.