Large-scale simulations of self-organizing neural networks on parallel computers: application to biological modelling

Abstract In this contribution we report about a study of a very versatile neural network algorithm known as “Self-organizing Feature Maps” and based on earlier work of Kohonen [1,2]. In its original version, the algorithm addresses a fundamental issue of brain organization, namely how topographically ordered maps of sensory information can be formed by learning. This algorithm is investigated for a large number of neurons (up to 16 K) and for an input space of dimension d⩽900. To meet the computational demands this algorithm was implemented on two parallel machines, on a self-built Transputer systolic ring and on a Connection Machine CM-2. We will present below 1. (i) a simulation based on the feature map algorithm modelling part of the synaptic organization in the “hand-region” of the somatosensory cortex, 2. (ii) a study of the influence of the dimension of the input-space on the learning process, 3. (iii) a simulation of the extended algorithm, which explicitly includes lateral interactions, and 4. (iv) a comparison of the transputer-based “coarse-grained” implementation of the model, and the “fine-grained” implementation of the same system on the Connection Machine.

[1]  Helmut Grubmüller,et al.  Molecular dynamics simulation on a parallel computer. , 1990 .

[2]  Roman Bek,et al.  Discourse on one way in which a quantum-mechanics language on the classical logical base can be built up , 1978, Kybernetika.

[3]  G. Blasdel,et al.  Voltage-sensitive dyes reveal a modular organization in monkey striate cortex , 1986, Nature.

[4]  T. M. Martinetz,et al.  3D neural net for learning visuomotor-coordination of a robot arm , 1989, International 1989 Joint Conference on Neural Networks.

[5]  Klaus Schulten,et al.  Topology-conserving maps for learning visuo-motor-coordination , 1989, Neural Networks.

[6]  G. J. Hueter.  Solution of the traveling salesman problem with an adaptive ring , 1988, IEEE 1988 International Conference on Neural Networks.

[7]  Teuvo Kohonen,et al.  Speech recognition based on topology-preserving neural maps , 1989 .

[8]  Helge Ritter.  Asymptotic level density for a class of vector quantization processes , 1991, IEEE Trans. Neural Networks.

[9]  T. Kohonen.  Self-organized formation of topographically correct feature maps , 1982 .

[10]  W. Daniel Hillis,et al.  The connection machine , 1985 .

[11]  C. Malsburg,et al.  How to label nerve cells so that they can interconnect in an ordered fashion. , 1977, Proceedings of the National Academy of Sciences of the United States of America.

[12]  N Suga,et al.  Disproportionate tonotopic representation for processing CF-FM sonar signals in the mustache bat auditory cortex. , 1976, Science.

[13]  T. Kohonen,et al.  Representation of sensory information in self-organizing feature maps , 1987 .

[14]  J. Pearson,et al.  Plasticity in the organization of adult cerebral cortical maps: a computer simulation based on neuronal group selection , 1987, The Journal of neuroscience : the official journal of the Society for Neuroscience.

[15]  Helge J. Ritter,et al.  A neural network model for the formation of topographic maps in the CNS: development of receptive fields , 1990, 1990 IJCNN International Joint Conference on Neural Networks.

[16]  Klaus Schulten,et al.  Large-Scale Simulation of a Self-organizing Neural Network: Formation of a Somatotopic Map , 1989 .

[17]  D. J. Felleman,et al.  Progression of change following median nerve section in the cortical representation of the hand in areas 3b and 1 in adult owl and squirrel monkeys , 1983, Neuroscience.

[18]  Nasser M. Nasrabadi,et al.  Vector quantization of images based upon the Kohonen self-organizing feature maps , 1988, ICNN.

[19]  M. Cynader,et al.  Somatosensory cortical map changes following digit amputation in adult monkeys , 1984, The Journal of comparative neurology.