One of the attractions of neural network approaches to pattern recognition is the use of a discrimination-based training method. We show that once we have modified the output layer of a multilayer perceptron to provide mathematically correct probability distributions, and replaced the usual squared error criterion with a probability-based score, the result is equivalent to Maximum Mutual Information training, which has been used successfully to improve the performance of hidden Markov models for speech recognition. If the network is specially constructed to perform the recognition computations of a given kind of stochastic model based classifier then we obtain a method for discrimination-based training of the parameters of the models. Examples include an HMM-based word discriminator, which we call an'Alphanet'.

With the growing interest in the practical use of neural networks, addressing the problem of customiling networks for specific applications is becoming increasingly critical. It has repeatedly been observed that different network structures and learning parameters can substantially affect performance.

In this paper, several techniques for addressing this problem are discussed in the context of an isolated word recognition task.

A methodology for faster supervised learning in dynamical nonlinear neural networks is presented. It exploits the concept of adjoint operntors to enable computation of changes in the network's response due to perturbations in all system parameters, using the solution of a single set of appropriately constructed linear equations. The lower bound on speedup per learning iteration over conventional methods for calculating the neuromorphic energy gradient is O(N2), where N is the number of neurons in the network. 1 INTRODUCTION The biggest promise of artifcial neural networks as computational tools lies in the hope that they will enable fast processing and synthesis of complex information patterns. In particular, considerable efforts have recently been devoted to the formulation of efficent methodologies for learning (e.g., Rumelhart et al., 1986; Pineda, 1988; Pearlmutter, 1989; Williams and Zipser, 1989; Barhen, Gulati and Zak, 1989). The development of learning algorithms is generally based upon the minimization of a neuromorphic energy function. The fundamental requirement of such an approach is the computation of the gradient of this objective function with respect to the various parameters of the neural architecture, e.g., synaptic weights, neural Adjoint Operator Algorithms 499

We present an application of back-propagation networks to handwritten digit recognition. Minimal preprocessing of the data was required, but architecture of the network was highly constrained and specifically designed for the task. The input of the network consists of normalized images of isolated digits. The method has 1 % error rate and about a 9% reject rate on zipcode digits provided by the U.S. Postal Service. 1 INTRODUCTION The main point of this paper is to show that large back-propagation (BP) networks can be applied to real image-recognition problems without a large, complex preprocessing stage requiring detailed engineering. Unlike most previous work on the subject (Denker et al., 1989), the learning network is directly fed with images, rather than feature vectors, thus demonstrating the ability of BP networks to deal with large amounts of low level information. Previous work performed on simple digit images (Le Cun, 1989) showed that the architecture of the network strongly influences the network's generalization ability. Good generalization can only be obtained by designing a network architecture that contains a certain amount of a priori knowledge about the problem. The basic design principle is to minimize the number of free parameters that must be determined by the learning algorithm, without overly reducing the computational power of the network.

To achieve high-rate image data compression while maintainig a high quality reconstructed image, a good image model and an efficient way to represent the specific data of each image must be introduced. Based on the physiological knowledge of multi - channel characteristics and inhibitory interactions between them in the human visual system, a mathematically coherent parallel architecture for image data compression which utilizes the Markov random field Image model and interactions between a vast number of filter banks, is proposed.

We have developed graphics to visualize static and dynamic information in layered neural network learning systems. Emphasis was placed on creating new visuals that make use of spatial arrangements, size information, animation and color. We applied these tools to the study of back-propagation learning of simple Boolean predicates, and have obtained new insights into the dynamics of the learning process.

The goal in this work has been to identify the neuronal elements of the cortical column that are most likely to support the learning of nonlinear associative maps. We show that a particular style of network learning algorithm based on locally-tuned receptive fields maps naturally onto cortical hardware, and gives coherence to a variety of features of cortical anatomy, physiology, and biophysics whose relations to learning remain poorly understood.

It has been known for many years that specific regions of the working cerebral cortex display periodic variations in correlated cellular activity. While the olfactory system has been the focus of much of this work, similar behavior has recently been observed in primary visual cortex. We have developed models of both the olfactory and visual cortex which replicate the observed oscillatory properties of these networks. Using these models we have examined the dependence of oscillatory behavior on single cell properties and network architectures. We discuss the idea that the oscillatory events recorded from cerebral cortex may be intrinsic to the architecture of cerebral cortex as a whole, and that these rhythmic patterns may be important in coordinating neuronal activity during sensory processmg.

The mapping of the back-propagation and mean field theory learning algorithms onto a generic 2-D SIMD computer is described. This architecture proves to be very adequate for these applications since efficiencies close to the optimum can be attained. Expressions to find the learning rates are given and then particularized to the DAP array procesor.