This paper presents an analog current-mode VLSI implementation of an unsupervised clustering algorithm. The clustering algorithm is based on the popular ART1 algorithm [1], but has been modified resulting in a more VLSI-friendly algorithm [2], [3] that allows a more efficient hardware implementation with simple circuit operators, little memory requirements, modular chip assembly capability, and higher speed figures. The chip described in this paper implements a network that can cluster 100 binary pixels input patterns into up to 18 different categories. Modular expansibility of the system is directly possible by assembling an NM array of chips without any extra interfacing circuitry, so that the maximum number of clusters is 18M and the maximum number of bits of the input pattern is N100. Pattern classification and learning is performed in 1.8s, which is an equivalent computing power of 4.410 9 connections per second plus connection-updates per second. The chip has been fabricated in...

Neural networks are non-linear static or dynamical systems that learn to solve problems from examples. Those learning algorithms that require a lot of computing power could benefit from fast dedicated hardware. This paper presents an overview of digital systems to implement neural networks. We consider three options for implementing neural networks in digital systems: serial computers, parallel systems with standard digital components, and parallel systems with special-purpose digital devices. We describe many examples under each option, with an emphasis on commercially available systems. We discuss the trend toward more general architectures, we mention a few hybrid and analog systems that can complement digital systems, and we try to answer questions that came to our minds as prospective users of these systems. We conclude that support software and in general, system integration, is beginning to reach the level of versatility that many researchers will require. The next step appears ...

Signatures are used everyday to authorise the transfer of funds of millions of people. Bank checks, credit cards and legal documents all require our signatures. Forgeries in such transactions cost millions of dollars each year. By forgery is meant copying, falsifying, or altering any kind of written or printed matter for the purpose of defrauding others. Signature verification is the process carried out to determine whether a given signature is genuine or forged. This research investigated the suitability of using backpropagation neural networks for the task of off-line signature verification. Several network architectures were designed to investigate this task. All networks used for experiments were feedforward neural networks. Each of the designed networks was trained using the three backpropagation learning algorithms (Vanilla, Enhanced, and Batch). A number of experiments have been conducted to evaluate the performance of these architectures and training algorithms. The performance...

Signatures are used everyday to authorise the transfer of funds of millions of people. Bank checks, credit cards and legal documents all require our signatures. Forgeries in such transactions cost millions of dollars each year. By forgery is meant copying, falsifying, or altering any kind of written or printed matter for the purpose of defrauding others. Signature verification is the process carried out to determine whether a given signature is genuine or forged. There are two major methods of signature verification. One is an on-line method to measure the sequential data such as handwriting and pen pressure with a special device. The other is an off-line method that uses an optical scanner to obtain handwriting data written on paper. This research investigated the suitability of using multilayered feedforward neural networks for the task of off-line signature verification. The emphasis was on investigating the performance of the networks when presented with raw signature images and no...

The requirement for dense interconnect in artificial neural network systems has led researchers to seek high-density interconnect technologies. This paper reports an implementation using multi-chip modules (MCMs) as the interconnect medium. The specific system described is a self-organizing, parallel, and dynamic learning model which requires a dense interconnect technology for effective implementation; this requirement is fulfilled by exploiting MCM technology. The ideas presented in this paper regarding an MCM implementation of artificial neural networks are versatile and can be adapted to apply to other neural network and connectionist models. 1 Introduction Artificial neural networks offer an exciting area of research because of their ability to solve difficult problems, typically those dealing with pattern recognition. This ability is due, in part, to their densely interconnected parallel architecture. However, often neural networks are simulated on sequential computers and lose ...

this paper with slightly different emphases also appear in Intelligent Systems [Rud95c], and the International Journal of Artificial Intelligence Tools [Rud95e]