We introduce a framework for training architectures composed of several modules. This framework, which uses a statistical formulation of learning systems, provides a unique formalism for describing many classical connectionist algorithms as well as complex systems where several algorithms interact. It allows to design hybrid systems which combine the advantages of connectionist algorithms as well as other learning algorithms.

Through the use of neural network classifiers and careful feature selection, we have achieved high-accuracy speaker-independent spoken letter recognition. For isolated letters, a broad-category segmentation is performed Location of segment boundaries allows us to measure features at specific locations in the signal such as vowel onset, where important information resides. Letter classification is performed with a feed-forward neural network. Recognition accuracy on a test set of 30 speakers was 96%. Neural network classifiers are also used for pitch tracking and broad-category segmentation of letter strings.

M2/n' for sufficiently large values of M. Here, Poo(error) denotes the probability of error in the infinite sample limit, and is at most twice the error of a Bayes classifier. Although the value of the coefficient a depends upon the underlying probability distributions, the exponent of M is largely distribution free. We thus obtain a concise relation between a classifier's ability to generalize from a finite reference sample and the dimensionality of the feature space, as well as an analytic validation of Bellman's well known "curse of dimensionality." 1 INTRODUCTION One of the primary tasks assigned to neural networks is pattern classification. Common applications include recognition problems dealing with speech, handwritten characters, DNA sequences, military targets, and (in this conference) sexual identity. Two fundamental concepts associated with pattern classification are generalization (how well does a classifier respond to input data it has never encountered before?) and scalability (how are a classifier's processing and training requirements affected by increasing the number of features that describe the input patterns?).

People can capably tell if a human face is male or female. Recognizing the sex of conspecifics is important. ''''hile some animals use pheromones to recognize sex, in humans this task is primarily visual. How is sex recognized from faces? By and large we are unable to say. Although certain features are nearly pathognomonic for one sex or the other (facial hair for men, makeup or certain hairstyles for women), even in the absence of these cues the determination is made; and even in their presence, other cues may override. Sex-recognition in faces is thus a. prototypical pattern recognition task of the sort at which humans excel, but which has vexed traditional AI. It appea.rs to follow no simple algorithm, and indeed is modifiable according to fashion (makeup, hair etc).

The in previ us face recognition systems (Ka R· na e, 19;)y' ......--.-.. scaled to the range [0 -. 8]. Part of the training set and its d' tances between facial elements.

Static correlations among spike trains obtained from simulations of large arrays of cells are in agreement with the predictions from these Hamiltonians, and dynamic correlat.ions

We propose a new parallel-hierarchical neural network model to enable motor learning for simultaneous control of both trajectory and force.

Function approximation on high-dimensional spaces is often thwarted by a lack of sufficient data to adequately "fill" the space, or lack of sufficient computational resources. The technique of local variable selection provides a partial solution to these problems by attempting to approximate functions locally using fewer than the complete set of input dimensions.

While the network loading problem for 2-layer threshold nets is NPhard when learning from examples alone (as with backpropagation), (Baum, 91) has now proved that a learner can employ queries to evade the hidden unit credit assignment problem and PACload nets with up to four hidden units in polynomial time. Empirical tests show that the method can also learn far more complicated functions such as randomly generated networks with 200 hidden units. The algorithm easily approximates Wieland's 2-spirals function using a single layer of 50 hidden units, and requires only 30 minutes of CPU time to learn 200-bit parity to 99.7% accuracy.

We present a new neural network model for processing of temporal patterns. This model, the gamma neural model, is as general as a convolution delay model with arbitrary weight kernels w(t). We show that the gamma model can be formulated as a (partially prewired) additive model. A temporal hebbian learning rule is derived and we establish links to related existing models for temporal processing. 1 INTRODUCTION In this paper, we are concerned with developing neural nets with short term memory for processing of temporal patterns. In the literature, basically two ways have been reported to incorporate short-term memory in the neural system equations.