Eaton E. P. O. Biology University of Colorado Boulder, Co. 80309 Abstract We provide a computational description of the function of the Mauthner system.This is the brainstem circuit which initiates faststart escapes in teleost fish in response to sounds. Our simulations, usingbackpropagation in a realistically constrained feedforward network, have generated hypotheses which are directly interpretable interms of the activity of the auditory nerve fibers, the principle cells of the system and their associated inhibitory neurons. 1 INTRODUCTION 1.1 THE M.AUTHNER SYSTEM Much is known about the brainstem system that controls fast-start escapes in teleost fish. The most prominent feature of this network is the pair of large Mauthner cells whose axons cross the midline and descend down the spinal cord to synapse on primary motoneurons. The Mauthner system also includes inhibitory neurons, the PHP cells, which have a unique and intense field effect inhibition at the spikeinitiating zoneof the Mauthner cells (Faber and Korn, 1978). The Mauthner system is part of the full brainstem escape network which also includes two pairs of cells homologous to the Mauthner cell and other populations of reticulospinal neurons. With this network fish initiate escapes only from appropriate stimuli, turn away from the offending stimulus, and do so very rapidly with a latency around 15 msec in goldfish.

Critical features of a domain (such as invariance under translation, rotation, and permu- Clustering with a Domain-Specific Distance Measure 103 tation) are captured within the clustering procedure, rather than reflected in the properties of feature sets created prior to clustering. The distance measure and learning problem are formally described as nested objective functions. We derive an efficient algorithm by using optimization techniques that allow us to divide up the objective function into parts which may be minimized in distinct phases. The algorithm has accurately recreated 10 prototypes from a randomly generated sample database of 100 images consisting of 20 points each in 120 experiments. Finally, by incorporating permutation invariance in our distance measure, we have a technique that we may be able to apply to the clustering of graphs. Our goal is to develop measures which will enable the learning of objects with shape or structure. Acknowledgements This work has been supported by AFOSR grant F49620-92-J-0465 and ONR/DARPA grant N00014-92-J-4048.

We show how an "Elman" network architecture, constructed from recurrently connected oscillatory associative memory network modules, canemploy selective "attentional" control of synchronization to direct the flow of communication and computation within the architecture to solve a grammatical inference problem. Previously we have shown how the discrete time "Elman" network algorithm can be implemented in a network completely described by continuous ordinary differential equations. The time steps (machine cycles)of the system are implemented by rhythmic variation (clocking) of a bifurcation parameter. In this architecture, oscillation amplitudecodes the information content or activity of a module (unit), whereas phase and frequency are used to "softwire" the network. Only synchronized modules communicate by exchanging amplitudeinformation; the activity of non-resonating modules contributes incoherent crosstalk noise. Attentional control is modeled as a special subset of the hidden modules with ouputs which affect the resonant frequencies of other hidden modules. They control synchrony among the other modules anddirect the flow of computation (attention) to effect transitions betweentwo subgraphs of a thirteen state automaton which the system emulates to generate a Reber grammar. The internal crosstalk noise is used to drive the required random transitions of the automaton.

Given such a classification task in most cases it is not too difficult to devise a network architecture that is capable of learning the input-output relation as represented by a number of training examples.

The model consists of a two-dimensional sheet of leaky integrateand-fire neuronswith feedback connectivity consisting of local excitation andsurround inhibition. Each neuron is independently driven by homogeneous external noise. Spontaneous symmetry breaking occurs, resulting in the formation of "hotspots" of activity inthe network. These localized patterns of excitation appear as clusters that coalesce, disintegrate, or fluctuate in size while simultaneously movingin a random walk constrained by the interaction with other clusters. The emergent cross-correlation functions have a dual structure, with a sharp peak around zero on top of a much broader hill.

This paper describes the use of a convolutional neural network to perform address block location on machine-printed mail pieces. Locating the address block is a difficult object recognition problem because there is often a large amount of extraneous printing on a mail piece and because address blocks vary dramatically in size and shape. We used a convolutional locator network with four outputs, each trained to find a different corner of the address block. A simple set of rules was used to generate ABL candidates from the network output. The system performs very well: when allowed five guesses, the network will tightly bound the address delivery information in 98.2% of the cases. 1 INTRODUCTION The U.S. Postal Service delivers about 350 million mail pieces a day.

We propose a new distance measure which is optimal for the task of local PCA. Our results with speech and image data indicate that the nonlinear techniques provide more accurate encodings than PCA. Our local linear algorithm produces more accurate encodings (except for one simulation with image data), and trains much faster than five layer auto-associative networks. Acknowledgments This work was supported by grants from the Air Force Office of Scientific Research (F49620-93-1-0253) and Electric Power Research Institute (RP8015-2). The authors are grateful to Gary Cottrell and David DeMers for providing their image database and clarifying their experimental results. We also thank our colleagues in the Center for Spoken Language Understanding at OGI for providing speech data.

The conventional Bayesian justification of backprop is that it finds the MAP weight vector. As this paper shows, to find the MAP io function instead one must add a correction tenn to backprop. That tenn biases one towards io functions with small description lengths, and in particular favors (somekinds of) feature-selection, pruning, and weight-sharing.

Signal processing and classification algorithms often have limited applicability resulting from an inaccurate model of the signal's underlying structure.We present here an efficient, Bayesian algorithm for modeling a signal composed of the superposition of brief, Poisson-distributed functions. This methodology is applied to the specific problem of modeling and classifying extracellular neural waveforms which are composed of a superposition of an unknown number of action potentials CAPs). Previous approaches have had limited success due largely to the problems of determining the spike shapes, deciding how many are shapes distinct, and decomposing overlapping APs. A Bayesian solution to each of these problems is obtained by inferring a probabilistic model of the waveform. This approach quantifies the uncertainty of the form and number of the inferred AP shapes and is used to obtain an efficient method for decomposing complex overlaps. This algorithm can extract many times more information than previous methods and facilitates the extracellular investigation of neuronal classes and of interactions within neuronal circuits.

Mozer, 1991; Bharucha, 1992), the models rarely explain the way in which musical representations are constructed and learned.