[Proceedings] 1991 IEEE International Joint Conference on Neural Networks

18-21 Nov. 1991

Filter Results

Displaying Results 1 - 25 of 444
  • Operational fault tolerance of the ADAM neural network system

    Publication Year: 1991, Page(s):1 - 6 vol.1
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (357 KB)

    The author investigates the fault tolerance of the Advanced Distributed Associative Memory (ADAM), focusing on its operational use. The effect of the reliability of recall of variously configured ADAM systems is examined by injecting faults individually, and also in various combinations since correlations between them will influence their overall effect on the system. Analysis of the results indic... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A general purpose neurocomputer

    Publication Year: 1991, Page(s):361 - 366 vol.1
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (252 KB)

    Presents a neural network, composed of linear units with threshold, as the CPU of a stored program MIMD architecture. The Caianiello formalism, is introduced as an aid to implement the arithmetic and control algorithms, needed for the smooth running of this general-purpose system. That is, in the neural net both the arithmetic and logic algorithms and the operating system have been implemented. Th... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Building up neuromimetic machines with LNeuro 1.0

    Publication Year: 1991, Page(s):602 - 607 vol.1
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (346 KB)

    The state of experiments on neural networks simulations on a parallel architecture is presented. The computing device, LNeuro 1.0, is based on an existing coarse-grain parallel framework (INMOS Transputers), improved with finer grain parallel abilities through VLSI modules. A digital architecture, scalable and flexible enough to be useful for simulating various kinds of networks and paradigms, was... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • 1991 IEEE International Joint Conference on Neural Networks (Cat. No.91CH3065-0)

    Publication Year: 1991
    Request permission for commercial reuse | PDF file iconPDF (47 KB)
    Freely Available from IEEE
  • Nonlinear neural field filters

    Publication Year: 1991, Page(s):1885 - 1890 vol.3
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (132 KB)

    Design, stability and implementation of nonlinear neural field filters are examined. The input and output of the neural field filters are vector fields. A neural transform is used to represent the input, output signals and the transfer function of the neural field filter. It is concluded that the Lyapunov conditions for such fields are taken care of by a novel extension of the Routh stability crit... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Approximations of mappings and application to translational invariant networks

    Publication Year: 1991, Page(s):2294 - 2298 vol.3
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (209 KB)

    The author studies the approximation of continuous mappings and dichotomies by one-hidden-layer networks, from a computational point of view. The approach is based on a new approximation method, specially designed for constructing small networks. Upper bounds are given on the size of these networks. These results are specialized to the case of transitional invariant networks, i.e., networks whose ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Training layered perceptrons using low accuracy computation

    Publication Year: 1991, Page(s):554 - 559 vol.1
    Cited by:  Papers (7)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (208 KB)

    It is demonstrated that the random search approach to training layered perceptrons can be performed using low-accuracy computational precision, and therefore can be implemented using analog computational accuracy. In spite of their numerical stability, random search techniques suffer from ever-increasing search time as dimensionality grows. In response, the authors introduce a modified random sear... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Discovering production rules with higher order neural networks: a case study. II

    Publication Year: 1991, Page(s):547 - 553 vol.1
    Cited by:  Papers (11)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (332 KB)

    It is demonstrated by example that neural networks can be used successfully for automatic extraction of production rules from empirical data. The case considered is a popular public domain database of 8124 mushrooms. With the use of a term selection algorithm, a number of very accurate mask perceptrons (a kind of high-order network or polynomial classifier) have been developed. Then rounding of sy... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Kohonen's feature maps applied to ordered clustering applications

    Publication Year: 1991, Page(s):596 - 601 vol.1
    Cited by:  Papers (1)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (224 KB)

    An application to a computer-aided design database using Kohonen's self-organizing feature maps is discussed. An input representation is chosen to decode the input geometric drawings into a vector of 12 elements for each drawing. The input vectors are then mapped to 1D Kohonen feature maps. Upon convergence, the output nodes organize themselves in ascending or descending order, resulting in a sort... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Joint VAr controller implemented in an artificial neural network environment

    Publication Year: 1991, Page(s):1336 - 1342 vol.2
    Cited by:  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (304 KB)

    Artificial neural network software has been trained to perform the joint VAr control function and demonstrated on three large synchronous condensers. Training rate depends on the nature of the training data set. It is noted that training data sets should be designed to place the majority of the data in the areas where the transfer function is the most critical View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Multifunctional layered network with quantizer neurons

    Publication Year: 1991, Page(s):541 - 546 vol.1
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (272 KB)

    The authors propose a multifunctional layered network (MFLN) with a quantizer neuron model and describe the principles of the quantizer neuron and the structure of the network for a character recognition system. Each layer of the MFLN has a specific function defined by the quantizer input of the quantizer neuron, and its learning speed is extremely fast. The authors have applied it to a character ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An electronic artificial neuron with biologically realistic behavior

    Publication Year: 1991, Page(s):2490 - 2495 vol.3
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (296 KB)

    An electronic artificial neuron which exhibits dynamic behavior that closely follows the observed stimulus-response characteristics of single isolated neurons is described. The artificial neuron has been implemented using analog and digital circuitry and has been designed to permit relatively easy alteration of its dynamic and static behavior. The primary purpose of the artificial neuron is to pro... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A neural network correlator for satellite imagery and ground truth data in a geographical information system

    Publication Year: 1991, Page(s):590 - 595 vol.1
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (332 KB)

    The authors describe results of research on using neural networks as a computational tool to capture the correlation between ground truth data and corresponding SPOT satellite imagery, and then incorporating the correlator within GRASS (Geographic Resources Analysis Support System), a geographical information system (GIS), to transform the whole imagery into land condition maps covering an entire ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Hybrid calibration of CCD cameras using artificial neural nets

    Publication Year: 1991, Page(s):337 - 342 vol.1
    Cited by:  Papers (11)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (264 KB)

    The authors first discuss the physical and mathematical model of CCD (charge coupled device) cameras on which the standard photogrammetric calibration of the cameras is based. Then they introduce artificial neural networks in order to improve the classical calibration of the CCD cameras, and thus develop a new method to calibrate CCD cameras. In this set-up, a feedforward artificial neural network... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Capacity of recurrent networks

    Publication Year: 1991, Page(s):1330 - 1335 vol.2
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (204 KB)

    It is shown that one can assign n arbitrarily linearly independent target outputs as fixed points for a recurrent network and that this assignment is independent of the dynamics of the system. The recurrent network has a capacity of at least n, which is the size of the net. The numerical complexity for obtaining the weights is discussed View full abstract»

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

    Publication Year: 1991, Page(s):36 - 41 vol.1
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (252 KB)

    In distributed representations, an object is represented as a pattern of activation over a number of units. Not all patterns of activity are valid in a given context. In the case of the Hopfield model of content addressable memory only the patterns of activation corresponding to the stored memories are valid. The problem of validating a given pattern of activity is called the validity problem. It ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Genetic algorithms: hints from biological science

    Publication Year: 1991, Page(s):426 - 429 vol.1
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (184 KB)

    C.T. Walbridge (Technology Review, vol.92, p.46-53, (1989)) discussed how ideas from evolution theory can be used to help in solving problems often addressed by the traditional symbolic methods of artificial intelligence. Choosing VLSI design as an example, he suggested that a chip chromosome would evolve just as genes evolve and the surviving organisms would mate to produce offspring having the c... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Multiple training concept for back-propagation networks

    Publication Year: 1991, Page(s):535 - 540 vol.1
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (188 KB)

    The multiple training concept first applied to bidirectional associative memory training is applied to the one-sweep back-propagation algorithm. The algorithm is called multiple training back-propagation. Computer simulations show that by putting different weights on different pairs in the energy function, this algorithm can increase the training speed of the network. The pair weights are updated ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Mapping multi-layer attributed graphs onto recognition network

    Publication Year: 1991, Page(s):1470 - 1475 vol.2
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (288 KB)

    A methodology of synthesizing a neocognitron is presented. The goal is that the system parameters is a neocognitron can be `programmed' rather than learned through laborious training. The tool used is the attribute graph theory. Using a set of attribute graphs describing structural and contextual information of different classes of patterns, one can synthesize a neocognitron through a mapping algo... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Towards a unified hybrid model of category representation

    Publication Year: 1991, Page(s):2484 - 2489 vol.3
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (356 KB)

    The author (1991) has argued that existing symbolic and subsymbolic models of category representation fail to capture critical features of hierarchy and partonomy. In this paper a hybrid connectionist model that addresses these shortfalls is described. By way of demonstration, it is shown how it can deliver faster basic level categorization times in categorizing isolated objects, and equality of b... View full abstract»

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

    Publication Year: 1991, Page(s):128 - 133 vol.1
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (284 KB)

    The high degree complexity of the features associated with a unit in neural networks suggested that the introduction of fuzziness into the activity of the unit would be appropriate. It is demonstrated that the idea of imprecise distinction between excitation and inhibition can be manipulated easily by fuzzy activation functions. A mathematical formulation of fuzzy activation functions, which are g... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Some experiments on the use of genetic algorithms in a Boltzmann machine

    Publication Year: 1991, Page(s):2645 - 2652 vol.3
    Cited by:  Papers (1)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (452 KB)

    The authors combined a genetic algorithm (GA) and simulated annealing to form a genetic Boltzmann machine (GBM) and attempted to understand the properties of such an architecture by experiments. Results of other experiments are also shown relating to the selection of parameters for the GA. The effects of population, different crossover point operators, and hidden units are illustrated. It is concl... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The abilities of neural networks to abstract and to use abstractions

    Publication Year: 1991, Page(s):584 - 589 vol.1
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (332 KB)

    The representation of concrete objects in neural networks is one of the important research areas in cognitive modeling. In the present work, the problem of the degree of abstractness is briefly considered, and some central concepts involving the representation of abstract entities are discussed. The epistemological status of the ability to use abstractions is discussed, and a proposal concerning t... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A lexisearch algorithm for traveling salesman problem

    Publication Year: 1991, Page(s):2521 - 2527 vol.3
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (276 KB)

    A data-guided lexisearch algorithm for the traveling salesman problem (TSP) is presented along with an illustrative example. The algorithm is a modification of the lexisearch approach to TSP as proposed by S.N.N. Pandit (1962). By utilizing the information provided by appropriate statistics computed from the cost data of the TSP, the nodes of the network cities are renamed and an alphabet table is... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fuzzy approach to a neural network

    Publication Year: 1991, Page(s):1675 - 1680 vol.2
    Cited by:  Patents (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (240 KB)

    A mathematical model of a multilevel, multilayer neural network has been developed which uses fuzzy theory for classification. The developed architecture allows imprecise or incomplete knowledge gained at one level of the network to be forwarded as input to the next level. The partial knowledge gained, which can be viewed as a set of fuzzy decisions, is mathematically represented in the network co... View full abstract»

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