The human brain is far superior to a modern computer in its ability to do associative recall. Many theorists believe that one of the important functions of primate neocortex is "associative memory". Palm’s network [1] is one of the most powerful associative memory models available. To study variations of this basic model, we have built a multiprocessor based Palm simulator that executes on our ...

Traditionally the parametric grammar based approach to the modeling and recognition of temporal sequences using hidden Markov models HMMs involves a very crucial step which requires human experts to determine a priori the appropriate model architecture to work on This includes among other things determining the number of states in the proba bilistic grammar and the probabilistic transitions bet...

The paper gives a simple algorithm that allows us to eliminate correlation between input binary patterns by changing their dimensionality. A neural network that is a variant of vector associative memory is used to recognize redimensioned patterns. Having capacity much greater than conventional neural networks, the resulting associative memory can recognize highly noisy and correlated input patt...

We propose to present a novel syntactic pattern matching technique [1] that combines the correlative learning and generalisation properties of associative memories, with the parallel and distributed operation of cellular automata [2]. The tool is used for recognizing two dimensional objects in images. Each section of the object to be studied is represented by a cell, initially containing a low ...

This paper introduces the sparse auto-associative neural network (SAANN) in which the internal hidden layer output is forced to be sparse. This is achieved by adding a sparse regularization term to the original reconstruction error cost function, and updating the parameters of the network to minimize the overall cost. We show applicability of this network to phoneme recognition by extracting sp...

This paper describes current research on neural networks within the collaborative project "Vision by Associative Reasoning" 1. The major aim of the work is the matching of vector descriptions of maps to airbourne IR imagery. The primary application of the work is in the guidance of airbourne vehicles. The paper describes how neural networks are being used to identify urban and field areas and f...

Empirical bankruptcy prediction models have been proposed and widely used in the last decades or so. Historic solvent and default firm data are collected and labeled appropriately. Statistical and neural network models are then “trained” to fit these data. A major problem is the imbalance of data, i.e. much more solvent data than default data. We propose a auto-associative neural network(AANN) ...