By Howard C. Card (auth.), José G. Delgado-Frias, William R. Moore (eds.)
Neural community and synthetic intelligence algorithrns and computing have elevated not just in complexity but in addition within the variety of purposes. This in flip has posed an incredible want for a bigger computational energy that traditional scalar processors is probably not capable of convey successfully. those processors are orientated in the direction of numeric and knowledge manipulations. because of the neurocomputing standards (such as non-programming and studying) and the factitious intelligence specifications (such as symbolic manipulation and information illustration) a special set of constraints and calls for are imposed at the computing device architectures/organizations for those functions. learn and improvement of latest computing device architectures and VLSI circuits for neural networks and synthetic intelligence were elevated as a way to meet the hot functionality standards. This publication offers novel ways and tendencies on VLSI implementations of machines for those purposes. Papers were drawn from a couple of learn groups; the topics span analog and electronic VLSI layout, machine layout, machine architectures, neurocomputing and synthetic intelligence options. This publication has been geared up into 4 topic components that disguise the 2 significant different types of this e-book; the components are: analog circuits for neural networks, electronic implementations of neural networks, neural networks on multiprocessor platforms and functions, and VLSI machines for man made intelligence. the themes which are lined in each one sector are in short brought below.