NEUROCOMPUTING Neurocomputing
13 ( 1996) 93-94
Guest editorial
Presenting the special issue on Soft Computing This special issue of Neurocomputing contains a selection of papers originally presented at the 3rd International Conference on Fuzzy Logic, Neural Nets, and Soft Computing held August 1-7, 1994 in Iizuka, Japan. This was the third in a series of successful meetings held in Iizuka on the Kyushu island of southern Japan. The conference attracted a large and truly international crowd of researchers from twenty-one countries who met to discuss the new and exciting field of soft computing. The soft computing methodology traces its origins to biological and human-like information processing. The methodology requires, however, new theoretical formalisms around which novel efficient computing schemes may be organized. Neural networks, fuzzy and probabilistic logic, and genetic and evolutionary methods are presently among the most popular approaches which offer these formalisms. In addition to these quickly maturing methodologies, additional factors spur the growth of soft computing. Our times are characterized by two dominant phenomena: massive amounts of available data, and high (and steadily increasing) computational power of digital computers. Realizing this, scientists and engineers in the area of soft computing are beginning to capitalize on both rich information environments and computing resources. By harnessing the paradigms of learning from data, and including approximate reasoning and evolutionary approaches, scientists working in the area of soft computing are working on responses to questions that are hard to answer otherwise. During the last ten years, renewed research interest in neural networks, fuzzy and evolutionary computing methods has brought numerous improvements to the area of information processing and intelligent systems. In almost every area of technology, science and business, advances and applications of soft computing have provided new results. In many cases, these methods have provided efficient and often unique solutions for tasks that would otherwise require specialized algorithms or their fusion, and overly long development time. We focus in this special issue on the neurocomputing-oriented track of the conference for the benefit of those who were unable to attend. All papers are extended and re-reviewed versions of the best among presented ones. Contributions range from the general and theoretical aspects of neural networks to applications-oriented papers. The scope of applications ranges from texture analysis, speech recognition, and anomaly detection to control, navigation, design, and expert systems. We would like to thank the Editor-in-Chief for the opportunity to organize this special issue. 0925-23 12/%/$15.00 Q 1996 Elwier PII SO925-23 12(96)00027-6
Science B.V. All rights resewed
94
Guest editorial/Neurocomputing
13 (19%) 93-94
We hope the readers will enjoy the issue and will attend one of the upcoming Iizuka conferences to share their views with others. Jacek M. Zurada Takeshi Yamakawa Guest Editors