Principles of Artificial Neural Networks

Artificial neural networks are most suitable for solving problems that are complex, ill-defined, highly nonlinear, of many and different variables, and/or stochastic. Such problems are abundant in medicine, in finance, in security and beyond.
This volume covers the basic theory and architecture of the major artificial neural networks. Uniquely, it presents 18 complete case studies of applications of neural networks in various fields, ranging from cell-shape classification to micro-trading in finance and to constellation recognition — all with their respective source codes. These case studies demonstrate to the readers in detail how such case studies are designed and executed and how their specific results are obtained.
The book is written for a one-semester graduate or senior-level undergraduate course on artificial neural networks. It is also intended to be a self-study and a reference text for scientists, engineers and for researchers in medicine, finance and data mining.
Contents:Introduction and Role of Artificial Neural NetworksFundamentals of Biological Neural NetworksBasic Principles of ANNs and Their Early StructuresThe PerceptronThe MadalineBack PropagationHopfield NetworksCounter PropagationLarge Scale Memory Storage and Retrieval (LAMSTAR) NetworkAdaptive Resonance TheoryThe Cognitron and the NeocognitronStatistical TrainingRecurrent (Time Cycling) Back Propagation NetworksReadership: Graduate and advanced senior students in artificial intelligence, pattern recognition & image analysis, neural networks, computational economics and finance, and biomedical engineering.