We present a neural model that can perform eye movements to a particular side of an object regardless of the position and orientation ofthe object in space, a generalization of a task which has been recently used by Olson and Gettner [4] to investigate the neural structureof object-centered representations. Our model uses an intermediate representation in which units have oculocentric receptive fields-just like collicular neurons-whose gain is modulated by the side of the object to which the movement is directed, as well as the orientation of the object. We show that these gain modulations are consistent with Olson and Gettner's single cell recordings in the supplementary eye field. This demonstrates that it is possible to perform an object-centered task without a representation involving anobject-centered map, viz., without neurons whose receptive fields are defined in object-centered coordinates. We also show that the same approach can account for object-centered neglect, a situation inwhich patients with a right parietal lesion neglect the left side of objects regardless of the orientation of the objects. Several authors have argued that tasks such as object recognition [3] and manipulation [4]are easier to perform if the object is represented in object-centered coordinates, arepresentation in which the subparts of the object are encoded with respect to a frame of reference centered on the object. Compelling evidence for the existence of such representations in the cortex comes from experiments on hemineglect-a neurological syndrome resulting from unilateral lesions of the parietal cortex such that a right lesion, for example, leads patients to ignore stimuli located on the left side of their egocentric space. Recently, Driver et al. (1994) showed that the deficit can also be object-centered.

The inverse of the Fisher information matrix is used in the natural gradientdescent algorithm to train single-layer and multi-layer perceptrons. We have discovered a new scheme to represent the Fisher information matrix of a stochastic multi-layer perceptron. Based on this scheme, we have designed an algorithm to compute the natural gradient. When the input dimension n is much larger than the number of hidden neurons, the complexity of this algorithm isof order O(n). It is confirmed by simulations that the natural gradient descent learning rule is not only efficient but also robust. 1 INTRODUCTION The inverse of the Fisher information matrix is required to find the Cramer-Rae lower bound to analyze the performance of an unbiased estimator. It is also needed in the natural gradient learning framework (Amari, 1997) to design statistically efficient algorithms for estimating parameters in general and for training neural networks in particular. In this paper, we assume a stochastic model for multilayer perceptrons.Considering a Riemannian parameter space in which the Fisher information matrix is a metric tensor, we apply the natural gradient learning rule to train single-layer and multi-layer perceptrons. The main difficulty encountered is to compute the inverse of the Fisher information matrix of large dimensions when the input dimension is high. By exploring the structure of the Fisher information matrix and its inverse, we design a fast algorithm with lower complexity to implement the natural gradient learning algorithm.

A rich body of data exists showing that recollection of specific information makesan important contribution to recognition memory, which is distinct from the contribution of familiarity, and is not adequately captured byexisting unitary memory models. Furthennore, neuropsychological evidence indicates that recollection is subserved by the hippocampus. We present a model, based largely on known features of hippocampal anatomy and physiology, that accounts for the following key characteristics ofrecollection: 1) false recollection is rare (i.e., participants rarely claim to recollect having studied nonstudied items), and 2) increasing interference leadsto less recollection but apparently does not compromise the quality of recollection (i.e., the extent to which recollected infonnation veridicallyreflects events that occurred at study).

Gabor filters are used as preprocessing stages for different tasks in machine vision and image processing. Their use has been partially motivated by findings that two dimensional Gabor filters can be used to model receptive fields of orientation selective neurons in the visual cortex (Daugman, 1980) and three dimensional spatiotemporal Gabor filters can be used to model biological image motion analysis (Adelson, 1985). A Gabor filter has a complex valued impulse response which is a complex exponential modulated by a Gaussian function.

In this paper, we present a novel hybrid architecture for continuous speech recognition systems. It consists of a continuous HMM system extended by an arbitrary neural network that is used as a preprocessor that takes several frames of the feature vector as input to produce more discriminative featurevectors with respect to the underlying HMM system. This hybrid system is an extension of a state-of-the-art continuous HMM system, andin fact, it is the first hybrid system that really is capable ofoutperforming thesestandard systems with respect to the recognition accuracy. Experimental results show an relative error reduction of about 10% that we achieved on a remarkably good recognition system based on continuous HMMsfor the Resource Management 1OOO-word continuous speech recognition task.

One of the current challenges to understanding neural information processing in biological systems is to decipher the "code" carried by large populations of neurons acting in parallel. We present an algorithm for automated discovery of stochastic firing patterns in large ensembles of neurons. The algorithm, from the "Helmholtz Machine" family, attempts to predict the observed spike patterns in the data. The model consists of an observable layer which is directly activated by the input spike patterns, and hidden units that are activated throughascending connections from the input layer. The hidden unit activity can be propagated down to the observable layer to create a prediction of the data pattern that produced it.

Support Vector (SV) Machines for pattern recognition, regression estimation and operator inversion exploit the idea of transforming into a high dimensional feature space where they perform a linear algorithm. Instead of evaluating this map explicitly, one uses Hilbert Schmidt Kernels k(x, y) which correspond to dot products of the mapped data in high dimensional space, i.e. k(x,y) ( I (x) · I (y)) (I) with I: .!Rn --*:F denoting the map into feature space. Mostly, this map and many of its properties are unknown. Even worse, so far no general rule was available.

Using displays of line orientations taken from Wolfe's experiments [1992], we study the hypothesis that the distinction between parallel versus serial processes arises from the availability of global information in the internal representations of the visual scene. The model operates in two phases. First, the visual displays are compressed via principal-component-analysis. Second, the compressed data is processed by a target detector module inorder to identify the existence of a target in the display. Our main finding is that targets in displays which were found experimentally tobe processed in parallel can be detected by the system, while targets in experimentally-serial displays cannot. This fundamental difference is explained via variance analysis of the compressed representations, providing a numerical criterion distinguishing parallelfrom serial displays. Our model yields a mapping of response-time slopes that is similar to Duncan and Humphreys's "search surface" [1989], providing an explicit formulation of their intuitive notion of feature similarity. It presents a neural realization ofthe processing that may underlie the classical metaphorical explanations of visual search.

In macaque inferotemporal cortex (IT), neurons have been found to respond selectivelyto complex shapes while showing broad tuning ("invariance") withrespect to stimulus transformations such as translation and scale changes and a limited tuning to rotation in depth.

The component gj's are usually relatively simple unimodal densities such as Gaussians. Density estimation with mixtures involves finding the locations, shapes, and weights of the component densities from the data (using for example the Expectation-Maximization (EM) procedure). Kernel density estimation canbe viewed as a special case of mixture modeling where a component is centered at each data point, given a weight of 1/N, and a common covariance structure (kernel shape) is estimated from the data. The quality of a particular probabilistic model can be evaluated by an appropriate scoring rule on independent out-of-sample data, such as the test set log-likelihood (also referred to as the log-scoring rule in the Bayesian literature).