By Topic

[Proceedings 1992] IJCNN International Joint Conference on Neural Networks

7-11 June 1992

Go

Filter Results

Displaying Results 1 - 25 of 138
  • A neural based approach of constraints satisfaction problem

    Publication Year: 1992, Page(s):588 - 593 vol.4
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (319 KB)

    The authors recall the constraint programming approach to solving a constraint-satisfaction problem, and show that for each problem which can be described by this approach a neural network can be designed which solves the problem. They give an application of this method for solving a task assignment problem and compare the results with those which are obtained by other approaches. Because this net... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neural net model of batch processes and optimization based on an extended genetic algorithm

    Publication Year: 1992, Page(s):519 - 524 vol.4
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (384 KB)

    The authors investigated the use of neural networks for modeling batch processes. A cascade neural network offered a solution from the experimental data which did not require the detailed knowledge of process kinetics. An extended genetic algorithm was adopted to generate the optimal trajectory for improving the desired process performance. The rule-inducer genetic algorithm is proposed for dynami... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Convergence and stability study of Hopfield's neural network for linear programming

    Publication Year: 1992, Page(s):525 - 531 vol.4
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (224 KB)

    Parameters that affect stability and convergence of the Hopfield model were identified by simulation. The Hopfield model used to solve optimization problems was defined by an analog electrical circuit. The authors illustrate that by introducing one additional amplifier a convergence with a good stability can be obtained. It is shown that convergence and stability can be obtained without oscillatio... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Real-time Chinese syllable recognition system with hierarchically structured neural network and transputer system

    Publication Year: 1992, Page(s):743 - 748 vol.4
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (192 KB)

    The authors propose a real-time Chinese syllable recognition system with a hierarchical neural network and a transputer system. The hierarchical neural network is composed of a type of classification network and three recognition networks. The Chinese syllable set is partitioned into a group of sub-sets. The classification network identifies the subset to which the input syllable belongs, and the ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Unsupervised and supervised data clustering with competitive neural networks

    Publication Year: 1992, Page(s):796 - 801 vol.4
    Cited by:  Papers (7)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (452 KB)

    The authors discuss objective functions for unsupervised and supervised data clustering and the respective competitive neural networks which implement these clustering algorithms. They propose a cost function for unsupervised and supervised data clustering which comprises distortion costs, complexity costs and supervision costs. A maximum entropy estimation of the clustering cost function yields a... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new approach to global optimization and its applications to neural networks

    Publication Year: 1992, Page(s):600 - 605 vol.4
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (404 KB)

    A new approach to global optimization that alternately rocks the landscape of the objective function and rolls the ball representing the current state of the variable down to the bottom of the nearest valley is presented. The degree of perturbation is determined by a parameter called rock level. The rock level decreases in the process. The ball gets rocked out of local minima and eventually settle... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Parallel mean field annealing neural network for solving traveling salesman problem

    Publication Year: 1992, Page(s):532 - 536 vol.4
    Cited by:  Papers (1)  |  Patents (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (220 KB)

    The authors propose a parallel mean field annealing (MFA) algorithm and a new energy function for finding traveling salesman optimal tours. The proposed parallel MFA neural network has the advantages of a simplified energy function, and that it converges more rapidly to an optimal solution. The experimental results showed that the parallel MFA and the new energy function can generate the optimal s... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Complementary aspects of topological maps and time delay neural networks for character recognition

    Publication Year: 1992, Page(s):444 - 449 vol.4
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (440 KB)

    A study comparing a back propagation-like network integrating feature selection notions introduced in neocognitron networks with a supervised learning algorithm, based on Kohonen's self-organizing feature maps, is presented. The two methods are applied to handwritten zipcode recognition. The results achieved by the networks are reported using common training and test sets. The complementary aspect... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Distinguishing line detection from texture segregation using a modular network-based model

    Publication Year: 1992, Page(s):392 - 397 vol.4
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (544 KB)

    An important early vision problem on how a bank of local spatial filters can be common to both line- and edge detection, and texture segregation is discussed. The authors introduce a network-based model for line- and edge detection and texture segregation. The network is based on the entropy driven artificial neural network (EDANN) model, a previously developed network module. Using a hierarchy of... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Learning visual coordinate transformations with competition

    Publication Year: 1992, Page(s):49 - 54 vol.4
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (508 KB)

    As the angle of gaze changes, so does the retinal location of the visual image of a stationary object. Since the object is correctly perceived as stationary, the retinopic coordinates of the object have been transformed into craniotopic coordinates somehow using eye position information. Neurons in area 7a of posterior parietal cortex in macaque monkeys are thought to contribute to this transforma... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A model of formal neural networks for unsupervised learning of binary temporal sequences

    Publication Year: 1992, Page(s):832 - 837 vol.4
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (312 KB)

    The authors propose a non-supervised model of formal neural networks to learn and recognize temporal sequences. Time is represented by its effect on processing and not as an additional dimension of inputs. Synaptic efficacy of a connection is the integration time of the signal passing through the connection. The only parameters subject to learning are connection integration times. It is assumed th... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On the training strategies of neural networks for speech recognition

    Publication Year: 1992, Page(s):749 - 754 vol.4
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (408 KB)

    The authors investigate how to introduce invariant features to speech recognition neural networks using conventional back propagation (BP), K-neighbor interpolation training (KNIT) with a number of time-shifted examples (TSEs) of the same training sample. The TSEs are employed for training of a multilayer perceptron (MLP) and a time-delay neural network (TDNN) structure to enrich the training samp... View full abstract»

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

    Publication Year: 1992, Page(s):767 - 772 vol.4
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (468 KB)

    Competitive Hebbian learning is extended to networks with trainable lateral connections, in addition to the trainable feedforward connections considered previously by the author (1991,1992). These recurrent systems are able to learn to respond to ordering in time of the input vectors. The theoretical framework for the extension of competitive Hebbian learning to recurrent systems is presented. Thi... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A self-organizing neural network for multidimensional approximation

    Publication Year: 1992, Page(s):802 - 807 vol.4
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (360 KB)

    A neural network based on the combination of a feature map (memory) and linear filters is proposed as a generalized adaptive processor for multidimensional nonlinear mapping. The self-organizing part of the system provides a progressively finer embedding of the input space as more units are added to the network. The linear filters, which tap from the memory, provide the function approximations. Le... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Sub-optimal solution screening in optimization by neural networks

    Publication Year: 1992, Page(s):606 - 611 vol.4
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (364 KB)

    The authors discuss a convergence condition of the Hopfield neural network to get the optimal or sub-optimal solutions of combinatorial optimization problems. For the TSP (traveling salesman problem), the condition to get its feasible solutions to coincide with the minimum points of the Hopfield neural network requires that the penalty parameter, which is the weight of a constraint function, must ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Genetically generated neural networks-I: representational effects

    Publication Year: 1992, Page(s):537 - 542 vol.4
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (296 KB)

    The author studied several applications of genetic algorithms (GAs) within the neural networks field. After generating a robust GA engine, the system was used to generate neural network circuit architectures. This was accomplished by using the GA to determine the weights in a fully interconnected network. The importance of the internal genetic representation was shown by testing different approach... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A neural computational scheme for extracting optical flow from the Gabor phase differences of successive images

    Publication Year: 1992, Page(s):450 - 456 vol.4
    Cited by:  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1236 KB)

    The authors propose a neurobiologically plausible representation of the Gabor phase information, and present a neural computation scheme for extracting visual motion information from the Gabor phase information. The scheme can compute visual motion accurately from a scene with illumination changes, while other neural schemes for optical flow must assume stable brightness. The computational tests o... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A neural network system model for active perception and invariant recognition of grey-level images

    Publication Year: 1992, Page(s):1 - 6 vol.4
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (504 KB)

    A method for parallel-sequential processing of gray-level images and their representation which is invariant to position, rotation, and scale has been developed. The method is based on the idea that an image is memorized and recognized by way of consecutive fixations of moving eyes on the most informative image fragments. The method provides the invariant representation of the image in each fixati... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Recognition of Japanese words by neural networks using vocal tract area

    Publication Year: 1992, Page(s):637 - 642 vol.4
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (288 KB)

    The authors present a new system for Japanese word recognition by neural networks using the vocal tract area. They present a method by which the vocal tract area is directly estimated from speech waves. The estimation method applies an adaptive inverse filter to the autocorrelation coefficients. A neural network learning algorithm developed by Y. Ishida et al. (1991), which is based on the conjuga... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Modeling neural network dynamics using iterative image reconstruction algorithms

    Publication Year: 1992, Page(s):398 - 402 vol.4
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (220 KB)

    Image reconstruction problems can be viewed as energy minimization problems and can be mapped onto a Hopfield neural network. For image reconstruction problems the authors describe the Gerchberg-Papoulis iterative method and the priorized discrete Fourier transform (PDFT) algorithm (C.L. Byrne et al., 1983). Both of these can be mapped onto a Hopfield neural network architecture, with the PDFT inc... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Biologically-based neural network model of color constancy and color contrast

    Publication Year: 1992, Page(s):55 - 60 vol.4
    Cited by:  Patents (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (376 KB)

    To determine the surface reflectance of an object independent of the illuminant, a system must use the spatiochromatic context of the image. The authors have developed a neural network model based on the anatomy and physiology of the visual projection from the retina to V4. The network combines color-opponent and contrast information to achieve a good degree of color constancy. This network has be... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A rapid learning orthonormal neural network for signal processing

    Publication Year: 1992, Page(s):265 - 270 vol.4
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (384 KB)

    The author describes a neural network architecture similar to the one suggested by Kolmogorov's existence theorem and a data processing method based on Fourier series. The resulting system, called the orthonormal neural network, can approximate any L2 mapping function between the input and output vectors without using hidden layers or the backpropagation rule. Because the trans... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An analog parallel-processing array for motion detection

    Publication Year: 1992, Page(s):327 - 332 vol.4
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (280 KB)

    The chip reported consists of an array of cells which functionally simulate the light-intensity-variation detecting neurons that exist in the retina of most highly developed animals. This chip performs motion detection using sub-nanoampere differentiation of the output currents of photodiodes arranged in a 2D array. As in most cases of image processing systems, the image projected onto the array m... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neurodynamical model of collective brain

    Publication Year: 1992, Page(s):115 - 121 vol.4
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (388 KB)

    A dynamical system which mimics collective purposeful activities of a set of units of intelligence is introduced and discussed. A global control of the unit activities is replaced by the probabilistic correlations between them. These correlations are learned during a long term period of performing collective tasks, and are stored in the synaptic interconnections. The model is represented by a syst... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Formant estimation from cepstral coefficients using a feedforward memoryless neural network

    Publication Year: 1992, Page(s):673 - 678 vol.4
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (376 KB)

    A method is described for estimating formants from cepstral coefficients using a memoryless feedforward neural network. The method was tested with vowel data from a large varied database. The neural network provided significantly better estimation of the formants than was possible with a linear transformation. However, the degradation in performance between training and test data suggests that the... View full abstract»

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