Typical methods for gradient descent in neural network learning involve calculation of derivatives based on a detailed knowledge of the network model. This requires extensive, time consuming calculations for each pattern presentation and high precision that makes it difficult to implement in VLSI. We present here a perturbation technique that measures, not calculates, the gradient. Since the technique uses the actual network as a measuring device, errors in modeling neuron activation and synaptic weights do not cause errors in gradient descent. The method is parallel in nature and easy to implement in VLSI. We describe the theory of such an algorithm, an analysis of its domain of applicability, some simulations using it and an outline of a hardware implementation.

The invariance of an objects' identity as it transformed over time provides a powerful cue for perceptual learning. We present an unsupervised learning procedure which maximizes the mutual information between the representations adopted by a feed-forward network at consecutive time steps. We demonstrate that the network can learn, entirely unsupervised, to classify an ensemble of several patterns by observing pattern trajectories, even though there are abrupt transitions from one object to another between trajectories. The same learning procedure should be widely applicable to a variety of perceptual learning tasks. 1 INTRODUCTION A promising approach to understanding human perception is to try to model its developmental stages. There is ample evidence that much of perception is learned.

How can artificial neural nets generalize better from fewer examples? In order to generalize successfully, neural network learning methods typically require large training data sets. We introduce a neural network learning method that generalizes rationally from many fewer data points, relying instead on prior knowledge encoded in previously learned neural networks. For example, in robot control learning tasks reported here, previously learned networks that model the effects of robot actions are used to guide subsequent learning of robot control functions. For each observed training example of the target function (e.g. the robot control policy), the learner explains the observed example in terms of its prior knowledge, then analyzes this explanation to infer additional information about the shape, or slope, of the target function. This shape knowledge is used to bias generalization when learning the target function. Results are presented applying this approach to a simulated robot task based on reinforcement learning.

We describe a model of visual word recognition that accounts for several aspects of the temporal processing of sequences of briefly presented words. The model utilizes a new representation for written words, based on dynamic time warping and multidimensional scaling. The visual input passes through cascaded perceptual, comparison, and detection stages. We describe how these dynamical processes can account for several aspects of word recognition, including repetition priming and repetition blindness.

A switching between apparently coherent (oscillatory) and stochastic episodes of activity has been observed in responses from cat and monkey visual cortex. We describe the dynamics of these phenomena in two parallel approaches, a phenomenological and a rather microscopic one. On the one hand we analyze neuronal responses in terms of a hidden state model (HSM). The parameters of this model are extracted directly from experimental spike trains. They characterize the underlying dynamics as well as the coupling of individual neurons to the network. This phenomenological model thus provides a new framework for the experimental analysis of network dynamics.

So far there has been no general method for relating extracellular electrophysiological measured activity of neurons in the associative cortex to underlying network or "cognitive" states. We propose to model such data using a multivariate Poisson Hidden Markov Model. We demonstrate the application of this approach for temporal segmentation of the firing patterns, and for characterization of the cortical responses to external stimuli. Using such a statistical model we can significantly discriminate two behavioral modes of the monkey, and characterize them by the different firing patterns, as well as by the level of coherency of their multi-unit firing activity. Our study utilized measurements carried out on behaving Rhesus monkeys by M. Abeles, E. Vaadia, and H. Bergman, of the Hadassa Medical School of the Hebrew University. 1 Introduction Hebb hypothesized in 1949 that the basic information processing unit in the cortex is a cell-assembly which may include thousands of cells in a highly interconnected network[l].

Representations for semantic information about words are necessary for many applications of neural networks in natural language processing. This paper describes an efficient, corpus-based method for inducing distributed semantic representations for a large number of words (50,000) from lexical coccurrence statistics by means of a large-scale linear regression. The representations are successfully applied to word sense disambiguation using a nearest neighbor method. 1 Introduction Many tasks in natural language processing require access to semantic information about lexical items and text segments.

Neural network models have been criticized for their inability to make use of compositional representations. In this paper, we describe a series of psychological phenomena that demonstrate the role of structured representations in cognition. These findings suggest that people compare relational representations via a process of structural alignment. This process will have to be captured by any model of cognition, symbolic or subsymbolic.

Networks with local inhibition are shown to have enhanced computational performance with respect to the classical Hopfield-like networks. In particular the critical capacity of the network is increased as well as its capability to store correlated patterns. Chaotic dynamic behaviour (exponentially long transients) of the devices indicates the overloading of the associative memory. An implementation based on a programmable logic device is here presented. A 16 neurons circuit is implemented whit a XILINK 4020 device.

The planar thallium-201 myocardial perfusion scintigram is a widely used diagnostic technique for detecting and estimating the risk of coronary artery disease. Neural networks learned to interpret 100 thallium scintigrams as determined by individual expert ratings. Standard error backpropagation was compared to standard LMS, and LMS combined with one layer of RBF units. Using the "leave-one-out" method, generalization was tested on all 100 cases. Training time was determined automatically from cross-validation perfonnance. Best perfonnance was attained by the RBF/LMS network with three hidden units per view and compares favorably with human experts.