Human genes are not continuous but rather consist of short coding regions (exons) interspersed with highly variable non-coding regions (introns). We apply HMMs to the problem of modeling exons, introns and detecting splice sites in the human genome. Our most interesting result so far is the detection of particular oscillatory patterns, with a minimal period ofroughly 10 nucleotides, that seem to be characteristic of exon regions and may have significant biological implications.

Previously, we had developed the concept of a Segmental Neural Net (SNN) for phonetic modeling in continuous speech recognition (CSR). This kind of neu ral network technology advanced the state-of-the-art of large-vocabulary CSR, which employs Hidden Marlcov Models (HMM), for the ARPA 1oo0-word Resource Management corpus. More Recently, we started porting the neural net system to a larger, more challenging corpus - the ARPA 20,Ooo-word Wall Street Journal (WSJ) corpus. During the porting, we explored the following research directions to refine the system: i) training context-dependent models with a regularization method; ii) training SNN with projection pursuit; and ii) combining different models into a hybrid system. When tested on both a development set and an independent test set, the resulting neural net system alone yielded a perfonnance at the level of the HMM system, and the hybrid SNN/HMM system achieved a consistent 10-15% word error reduction over the HMM system. This paper describes our hybrid system, with emphasis on the optimization methods employed.

Data clustering amounts to a combinatorial optimization problem to reduce the complexity of a data representation and to increase its precision. Central and pairwise data clustering are studied in the maximum entropy framework. For central clustering we derive a set of reestimation equations and a minimization procedure which yields an optimal number of clusters, their centers and their cluster probabilities. A meanfield approximation for pairwise clustering is used to estimate assignment probabilities. A se1fconsistent solution to multidimensional scaling and pairwise clustering is derived which yields an optimal embedding and clustering of data points in a d-dimensional Euclidian space. 1 Introduction A central problem in information processing is the reduction of the data complexity with minimal loss in precision to discard noise and to reveal basic structure of data sets. Data clustering addresses this tradeoff by optimizing a cost function which preserves the original data as complete as possible and which simultaneously favors prototypes with minimal complexity (Linde et aI., 1980; Gray, 1984; Chou et aI., 1989; Rose et ai., 1990). We discuss an objective function for the joint optimization of distortion errors and the complexity of a reduced data representation. A maximum entropy estimation of the cluster assignments yields a unifying framework for clustering algorithms with a number of different distortion and complexity measures. The close analogy of complexity optimized clustering with winner-take-all neural networks suggests a neural-like implementation resembling topological feature maps (see Figure 1).

A gradient descent algorithm for parameter estimation which is similar to those used for continuous-time recurrent neural networks was derived for Hodgkin-Huxley type neuron models. Using membrane potential trajectories as targets, the parameters (maximal conductances, thresholds and slopes of activation curves, time constants) were successfully estimated. The algorithm was applied to modeling slow non-spike oscillation of an identified neuron in the lobster stomatogastric ganglion. A model with three ionic currents was trained with experimental data. It revealed a novel role of A-current for slow oscillation below -50 mY. 1 INTRODUCTION Conductance-based neuron models, first formulated by Hodgkin and Huxley [10], are commonly used for describing biophysical mechanisms underlying neuronal behavior.

Hoever, the relation between our ability to perceive structure in stimuli, simulating 3-D objects, and neuronal properties has not been addressed up to date. Here we discuss a possibility, that area MT RF surrounds might be involved in shape-frommotion perception. We introduce a simplifying model of MT neurons and analyse the implications to SFM problem solving. 2 REDEFINING THE SFM PROBLEM 2.1 RELATIVE MOTION AS A CUE FOR RELATIVE DEPTH Since Helmholtz motion parallax is known to be a powerful cue providing information about both the structure of the surrounding environment and the direction of self-motion. On the other hand, moving objects also induce velocity fields allowing judgement about their shapes. We can capture both cases by assuming that an observer is tracking a point on a surface of interest.

A fast event-driven software simulator has been developed for simulating large networks of spiking neurons and synapses. The primitive network elements are designed to exhibit biologically realistic behaviors, such as spiking, refractoriness, adaptation, axonal delays, summation of post-synaptic current pulses, and tonic current inputs. The efficient event-driven representation allows large networks to be simulated in a fraction of the time that would be required for a full compartmental-model simulation. Corresponding analog CMOS VLSI circuit primitives have been designed and characterized, so that large-scale circuits may be simulated prior to fabrication. 1 Introduction Artificial neural networks typically use an abstraction of real neuron behaviour, in which the continuously varying mean firing rate of the neuron is presumed to carry the information about the neuron's time-varying state of excitation [1]. This useful simplification allows the neuron's state to be represented as a time-varying continuous-amplitude quantity.

Online cursive handwriting recognition is currently one of the most intriguing challenges in pattern recognition. This study presents a novel approach to this problem which is composed of two complementary phases. The first is dynamic encoding of the writing trajectory into a compact sequence of discrete motor control symbols. In this compact representation we largely remove the redundancy of the script, while preserving most of its intelligible components. In the second phase these control sequences are used to train adaptive probabilistic acyclic automata (PAA) for the important ingredients of the writing trajectories, e.g.

This paper proposes a practical optimization method for layered neural networks, by which the optimal model and parameter can be found simultaneously. 'i\Te modify the conventional information criterion into a differentiable function of parameters, and then, minimize it, while controlling it back to the ordinary form. Effectiveness of this method is discussed theoretically and experimentally.

Feed-forward networks of localized (e.g., Gaussian) units are an interesting alternative to the more frequently used networks of global (e.g., sigmoidal) units. It has been shown that with localized units one hidden layer suffices in principle to approximate any continuous function, whereas with sigmoidal units two layers are necessary. In the following we are considering radial basis function networks similar to those proposed by Moody & Darken (1989) or Poggio & Girosi (1990). Such networks consist of one layer L of Gaussian units.

Many well known neural network techniques for adaptive pattern classification and function approximation are inherently highly parallel, and thus have proven difficult to implement for real-time applications at a reasonable cost.