Scheduled System Maintenance
On Tuesday, September 26, IEEE Xplore will undergo scheduled maintenance from 1:00-4:00 PM ET.
During this time, there may be intermittent impact on performance. We apologize for any inconvenience.

IEEE 1988 International Conference on Neural Networks

24-27 July 1988

Filter Results

Displaying Results 1 - 25 of 161
  • IEEE International Conference on Neural Networks (IEEE Cat. No.88CH2632-8)

    Publication Year: 1988
    Request permission for commercial reuse | PDF file iconPDF (12 KB)
    Freely Available from IEEE
  • Asynchronous dynamics of random Boolean networks

    Publication Year: 1988, Page(s):1 - 7 vol.1
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (422 KB)

    A generalization of neural networks called Boolean networks is considered. Random (n,k)-networks consist of n processors, each connected randomly to k others, computing random k-input Boolean functions. The dynamic behavior of these networks has been studied extensively. The authors examine the asynchronous dynamics of these networks, and prove that convergence to fixpoints is assured for almost a... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Bifurcation theory methods for programming static or periodic attractors and their bifurcations in dynamic neural networks

    Publication Year: 1988, Page(s):9 - 16 vol.1
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (573 KB)

    Analytic methods of bifurcation theory are used to design algorithms for synthesis of analog neural networks with a precisely defined local vector field for pattern recognition that is created by a multiple bifurcation. For two memories thus far, static or oscillating spatial patterns can be stored, basin boundaries programmed, and the location of secondary bifurcations that introduce new attracto... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neuron layer with reciprocal inhibition as a mechanism of random choice

    Publication Year: 1988, Page(s):17 - 25 vol.1
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (366 KB)

    The author describes the dynamics of a neuron layer with reciprocal inhibition. In the presence of noise the layer works as a selector, performing random choice of a single neuron from the set of neurons with the maximum level of excitation. He discusses the following topics: two-basic-layer architectures with reciprocal inhibition; the Laplace transforms of the transients in the intervals between... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Stability theorem for lateral inhibition networks that is robust in the presence of circuit parasitics

    Publication Year: 1988, Page(s):27 - 36 vol.1
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (447 KB)

    In the analog VLSI implementation of neural systems, it is sometimes convenient to built lateral inhibition networks by using a locally connected on-chip resistive grid. A serious problem of unwanted spontaneous oscillation often arises with these circuits and renders them unusable in practice. The authors report a design approach that guarantees such a system will be stable, even though the value... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Dynamical properties of threshold automata with nearest-neighbor interactions on a regular lattice

    Publication Year: 1988, Page(s):37 - 43 vol.1
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (329 KB)

    It is demonstrated that threshold automata with quasirandom nearest-neighbor interactions on a regular two-dimensional lattice exhibit, by far, more frozen behavior than their infinite-range counterparts. Systems with units residing on a honeycomb lattice strictly remain in an ordered condition, whereas for the square lattice based on five nearest-neighbor interactions, a phase transition from ord... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An investigation on local minima of a Hopfield network for optimization circuits

    Publication Year: 1988, Page(s):45 - 51 vol.1
    Cited by:  Papers (12)  |  Patents (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (368 KB)

    The Hopfield architecture can be utilized in the VLSI implementation of several important optimization functions. A description is given of the properties of local minima in the energy function of Hopfield networks. A novel design technique to eliminate such local minima has been developed. The neural-based analog-to-digital converter is used as an example to demonstrate this design technique. Exp... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Recurrent nets for the storage of cyclic sequences

    Publication Year: 1988, Page(s):53 - 60 vol.1
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (385 KB)

    Analytic, iterative and constructive design strategies for binary discrete-time recursive nets with synchronous update are discussed. The authors find that first-order correlation nets can be designed to exhibit a number of stable resonance modes which act as cyclic attractors. These modes represent information encoded in the network and can be used to recognize noisy sequences.<> View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Statistical pattern recognition with neural networks: benchmarking studies

    Publication Year: 1988, Page(s):61 - 68 vol.1
    Cited by:  Papers (217)  |  Patents (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (613 KB)

    Three basic types of neural-like networks (backpropagation network, Boltzmann machine, and learning vector quantization), were applied to two representative artificial statistical pattern recognition tasks, each with varying dimensionality. The performance of each network's approach to solving the tasks was evaluated and compared, both to the performance of the other two networks and to the theore... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An improved elastic net method for the traveling salesman problem

    Publication Year: 1988, Page(s):69 - 76 vol.1
    Cited by:  Papers (22)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (439 KB)

    An elastic net method is presented for finding traveling-salesman tours; the method improves on the convergence properties of a recent model proposed by R. Durbin and D. Willshaw (1987). It combines the notion of symmetric forces described by the author in an earlier paper (1981) with an annealing schedule derived directly from the input data. The author demonstrates that the method converges abou... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A neural controller for collision-free movement of general robot manipulators

    Publication Year: 1988, Page(s):77 - 84 vol.1
    Cited by:  Papers (21)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (736 KB)

    An approach is presented for the collision-free control of general robot manipulators moving among a changing set of obstacles. The controller, based on a layered, neural network architecture, adapts to the specific eye/hand and arm/body kinematics of any arbitrarily shaped robot during an initial, unsupervised training phase. After training, the robot selects a target point by 'glancing' at it an... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Solution of the traveling salesman problem with an adaptive ring

    Publication Year: 1988, Page(s):85 - 92 vol.1
    Cited by:  Papers (21)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (423 KB)

    A ring of nodes adapts, using Kohonen learning, to form a complete circuit through N cities in the plane. The algorithm is simple because the topology of the solution is contained in the definition of the network and does not need to be imposed as an external constraint. A ten-city problem is used to demonstrate several variants of the basic algorithm which address different problems in the mappin... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Self-organising multilayer topographic mappings

    Publication Year: 1988, Page(s):93 - 100 vol.1
    Cited by:  Papers (7)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (468 KB)

    Minimization of distortion measures requires multilayer mappings to be topographic. The author shows this only for tree-like multilayer networks. He also shows how to modify the original topographic mapping learning algorithm to increase its convergence rate. A three-layer network can form linelike feature detectors which are just as good as those in a two-layer network. However, the author finds ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Why neural networks don't yet fly: inquiry into the neurodynamics of biological intelligence

    Publication Year: 1988, Page(s):1 - 7 vol.2
    Cited by:  Papers (8)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (491 KB)

    There are two documented hierarchical modes in which wetware brains sustain information for storage, transformation, and other operations. These modes are in some senses diametrically opposed but in other senses are strongly complementary. One is typified by the action potential and the point process; the other by the synaptic potential and the local mean field. In sensory systems the one is the b... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Race networks: a theory of competitive recognition networks based on the rate of reactivation of neurons in cortical columns

    Publication Year: 1988, Page(s):9 - 16 vol.2
    Cited by:  Papers (1)  |  Patents (9)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (649 KB)

    A theory of noniterative neural processing is presented for use in recognition networks performing competitive pattern classification. This scheme has advantages over the more iterative approaches in terms of processing speed, simplicity, and stability while learning. The theory derives from an examination of the operation of the simple cells of the primary visual cortex. A competitive pattern cla... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Vector quantization of images based upon the Kohonen self-organizing feature maps

    Publication Year: 1988, Page(s):101 - 108 vol.1
    Cited by:  Papers (129)  |  Patents (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (649 KB)

    A neural-network clustering algorithm proposed by T. Kohonen (1986, 88) is used to design a codebook for the vector quantization of images. This neural-network clustering algorithm, which is better known as the Kohonen self-organizing feature maps, is a two-dimensional set of extensively interconnected nodes or unit of processors. The synaptic strengths between the input and the output nodes repre... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Kohonen's self-organizing maps: exploring their computational capabilities

    Publication Year: 1988, Page(s):109 - 116
    Cited by:  Papers (51)  |  Patents (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (557 KB)

    The authors demonstrate that the computational capabilities of Kohonen's algorithm provide an unified approach to such diverse fields as sensory mappings, combinatorial optimization, and learning in motor control. For a discrete probability distribution of the training inputs, the formation of the mapping can be described as a probabilistic descent in a potential. In view of their wide applicabili... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adding a conscience to competitive learning

    Publication Year: 1988, Page(s):117 - 124 vol.1
    Cited by:  Papers (242)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (420 KB)

    There are a number of neural networks that self-organize on the basis of what has come to be known as Kohonen learning. The author introduces a modification of Kohonen learning that provides rapid convergence and improved representation of the input data. In many areas of pattern recognition, statistical analysis, and control, it is essential to form a nonparametric model of a probability density ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Invariant pattern recognition by means of fast synaptic plasticity

    Publication Year: 1988, Page(s):125 - 132 vol.1
    Cited by:  Patents (4)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (515 KB)

    A two-layer neural system for shift-invariant pattern recognition is proposed. Model neurons are endowed with physiological dynamics involving membrane potentials and axonic spikes. Synapses between the two layers are plastic and change according to spike coincidences (Hebbian rules). The first neural network (encoder network) extracts features from a presented pattern and codes the neighborhood r... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Conceptual graph knowledge systems as problem context for neural networks

    Publication Year: 1988, Page(s):133 - 140 vol.1
    Cited by:  Papers (2)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (616 KB)

    For a connectionist network to be able to learn to generalize well, there must be some correspondence between the structure/constraints of the net's architecture and those of the given problem space. Therefore, recourse to experiments with real-world problems will always be required in connectionist research. The author gives an outline of a problem area for which connectionist nets hold great pro... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Feedback stability and unsupervised learning

    Publication Year: 1988, Page(s):141 - 152 vol.1
    Cited by:  Papers (12)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (614 KB)

    Global stability is examined for nonlinear feedback dynamical systems subject to unsupervised learning. Only differentiable neural models are discussed. The unconditional stability of Hebbian learning systems is summarized in the adaptive bidirectional associative memory (ABAM) theorem. When no learning occurs, the resulting BAM models include Cohen-Grossberg autoassociators, Hopfield circuits, br... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • High-capacity exponential associative memories

    Publication Year: 1988, Page(s):153 - 160 vol.1
    Cited by:  Papers (23)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (415 KB)

    A generalized associative memory model with potentially high capacity is presented. A memory of this kind with M stored vectors of length N, can be implemented with M nonlinear neurons, N ordinary thresholding neurons, and 2MN binary synapses. It is shown that special cases of this model include the Hopfield and high-order correlation memories. A special case of the model, based on a neuron which ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Multilayer feedforward potential function network

    Publication Year: 1988, Page(s):161 - 171 vol.1
    Cited by:  Papers (20)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (600 KB)

    The authors present a multilayer feedforward network, called the Gaussian potential function network (GPFN), performing association or classification based on a set of potentially fields synthesized over the domain of input space by a number of Gaussian potential function units (GPFUs). A GPFU as a basic component of the GPFN is designed to generate a Gaussian form of a potential field. A weighted... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A formulation for the design of neural processors

    Publication Year: 1988, Page(s):173 - 180 vol.1
    Cited by:  Papers (16)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (429 KB)

    The author presents a mathematical framework for the design of neural networks as associative memories or as pattern classifiers. This framework takes into account the characteristics and the limitations of the electrical components which are the building blocks for the proposed devices. He also presents conditions that given vectors must satisfy to successfully store them in the network. The cond... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A BAM with increased information storage capacity

    Publication Year: 1988, Page(s):181 - 190 vol.1
    Cited by:  Papers (38)  |  Patents (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (684 KB)

    The commonly encountered version of B. Kosko's bidirectional associative memory (BAM) (1988) is generalized to include nonzero thresholds. This modified neural network is referred to as the nonhomogeneous BAM. It is shown that the nonhomogeneous BAM can have any number of stable states between 1 and 2/sup n/, where n is the lowest dimension of either layer of the BAM. Nonhomogeneous BAMs with 1,2,... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.