1998 IEEE International Joint Conference on Neural Networks Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98CH36227)

4-9 May 1998

Go

Filter Results

Displaying Results 1 - 25 of 158
  • The 1998 IEEE International Joint Conference on Neural Networks Proceedings [front matter]

    Publication Year: 1998, Page(s):i - xxxvi
    Request permission for commercial reuse | PDF file iconPDF (2194 KB)
    Freely Available from IEEE
  • Author's index

    Publication Year: 1998, Page(s): A
    Request permission for commercial reuse | PDF file iconPDF (466 KB)
    Freely Available from IEEE
  • A new random search method for neural network learning-RasID

    Publication Year: 1998, Page(s):2346 - 2351 vol.3
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (468 KB)

    This paper presents a novel random searching scheme called RasID for neural networks training. The idea is to introduce a sophisticated probability density function (PDF) for generating search vector. The PDF provides two parameters for realizing intensified search in the area where it is likely to find good solutions locally or diversified search in order to escape from a local minimum based on t... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Convergence properties of symmetric learning algorithm for pattern classification

    Publication Year: 1998, Page(s):2340 - 2345 vol.3
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (508 KB)

    The geometric learning algorithm (GLA) was proposed as an application of the affine projection algorithm (APA) for an adaptive filter to a perceptron. In the GLA, the connection weight vector w(n) is updated vertically towards the orthogonal complement of κ patterns. The GLA demonstrates some typical behavior when the learning rate λ is 2, which means that w(n) and w(n+1) are symmetri... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Robust nonlinear control using neural networks

    Publication Year: 1998, Page(s):2104 - 2109 vol.3
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (432 KB)

    In this article, the influence of uncertainty on weights and biases of neural networks on the input/output behavior is investigated. Moreover, a uncertainty description of uncertain neural networks is derived and an appropriate norm bound of the model uncertainty, which is needed for robust control design, is derived. Finally, feedback linearization is used in order to fully incorporate neural net... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Image recognition using fractal parameters

    Publication Year: 1998, Page(s):1883 - 1888 vol.3
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (352 KB)

    Concerns the applications of fractal theory to image recognition and we propose the method that can enhance learning rate and recognition rate by using fractal parameters that are composed of input vectors for a neural network in an image recognition model. Fractal parameters with the properties of self-similarity and recursiveness can recover lossless original images through iterating processes. ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A fast method for rule extraction in neural networks

    Publication Year: 1998, Page(s):2334 - 2339 vol.3
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (460 KB)

    Most methods for finding regularities or rule extraction based data start out from the idea that the representation of the data must evolve from a distributed representation of the information to a more localised representation which will represent the skeleton of the network. This idea involves the problem of needing long training times imposed by the backpropagation algorithm, as well as the err... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A robust neural controller for underwater robot manipulator

    Publication Year: 1998, Page(s):2098 - 2103 vol.3
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (416 KB)

    This paper presents a robust control scheme wing a multilayer neural network. The multilayer neural network acts as a compensator of the conventional sliding mode controller to maintain the control performance when the initial assumptions of uncertainty bounds are not valid. The proposed controller applies to control the robot manipulator operating under the sea which has large uncertainties such ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Phased backpropagation: a hybrid of BPTT and temporal BP

    Publication Year: 1998, Page(s):2262 - 2267 vol.3
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (420 KB)

    We present a synthesis of backpropagation through time (BPTT) and temporal backpropagation (TB) that permits the efficiency of TB in dealing with delay lines to be combined with the generality of BPTT with respect to arbitrary discrete-time network structures. We express our formulation in terms of an ordered network, subsuming less general network architectures such as time-delay networks and rec... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Motion segmentation using temporal block matching and LEGION

    Publication Year: 1998, Page(s):2069 - 2074 vol.3
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (504 KB)

    A motion segmentation method is proposed for an input sequence of random dot and binary images. The method is composed of two main stages, inspired by primate visual system. The first stage determines local velocity information at each location in every image frame using its two neighboring image frames. Measurements of a particular velocity at all locations form the corresponding velocity layer. ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Scene segmentation in video sequences by an RPCL neural network

    Publication Year: 1998, Page(s):1877 - 1882 vol.3
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (396 KB)

    Video database management systems require efficient methods to abstract video information. Identification of shots in a video sequence is an important task for summarizing the content of a video. We describe a neural network based technique for automatic clustering of video frames in video sequences. From each frame the features that describe the image content are extracted to form a signature. Th... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A neural based segmentation and recognition technique for handwritten words

    Publication Year: 1998, Page(s):1738 - 1742 vol.3
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (508 KB)

    Artificial neural networks (ANNs) have been successfully applied to optical character recognition (OCR) yielding excellent results. In this paper a technique is presented that segments difficult printed and cursive handwriting, and then classifies the segmented characters. A conventional algorithm is used for the initial segmentation of the words, while an ANN is used to verify whether an accurate... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Designing regularizers by minimizing generalization errors

    Publication Year: 1998, Page(s):2328 - 2333 vol.3
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (432 KB)

    To improve generalization ability, a regularizer is frequently used. An approach proposed here is to regard the estimate of model parameters as a function of those without a regularizer. By minimizing the calculated generalization error, the optimal function parameters and model parameters can be obtained. In the paper linear regression is adopted to carry out theoretical computation of generaliza... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Noise suppression in training data for improving generalization

    Publication Year: 1998, Page(s):2236 - 2241 vol.3
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (420 KB)

    Multilayer feedforward neural networks are trained using the error backpropagation (BP) algorithm. This algorithm minimizes the error between outputs of a neural network (NN) and training data. Hence, in the case of noisy training data, a trained network memorizes noisy outputs for given inputs. Such learning is called rote memorization learning (RML). In this paper we propose error correcting mem... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A computational model of avoidance behavior

    Publication Year: 1998, Page(s):2092 - 2097 vol.3
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (472 KB)

    Learned avoidance behavior is critical to animal survival but has proven difficult for animal learning theorists to model. The authors propose a computational model of the highest layer of an hierarchical control system responsible for learning and behavior. The proposed layer consists of a network of associative control processes (ACPs) that employ the drive-reinforcement learning mechanism. A ne... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Regularization effect of weight initialization in back propagation networks

    Publication Year: 1998, Page(s):2258 - 2261 vol.3
    Cited by:  Papers (2)  |  Patents (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (264 KB)

    Complexity control of a learning method is critical for obtaining good generalization with finite training data. We discuss complexity control in multilayer perceptron (MLP) networks trained via backpropagation. For such networks, the number of hidden units and/or network weights is usually used as a complexity parameter. However, application of backpropagation training introduces additional mecha... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Tip-position tracking for flexible-link manipulators using artificial neural networks: experimental results

    Publication Year: 1998, Page(s):2063 - 2068 vol.3
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (484 KB)

    Presents experimental evaluation of the performance of neural network-based controllers for tip position tracking of flexible-link manipulators. Four different neural network schemes are proposed based on the output re-definition approach. The new output is defined assuming no a priori knowledge about the payload mass. The first two schemes are developed using a modified version of the “feed... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Self-organization, scaling, and parallelism

    Publication Year: 1998, Page(s):2039 - 2044 vol.3
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (884 KB)

    The problem of learning in the absence of external intelligence is discussed in the context of a simple model. The model departs from the traditional gradient-descent based approaches to learning by operating at a highly susceptible “critical” state, with low activity and sparse connections between firing neurons. Quantitative studies in the context of two simple association tasks demo... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An adaptive robust PCA neural network

    Publication Year: 1998, Page(s):2288 - 2293 vol.3
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (436 KB)

    We find one way to improve the robustness of principal component analysis (PCA) based on a reconstruction error model. First, we discuss and compare the methods to analyze the robustness of the PCA algorithm. A new adaptive algorithm of robust PCA based on the structure of a single-layer neural network (NN) is developed with the modification of the cost function which can be acquired through model... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Identifying part-of-speech patterns for automatic tagging

    Publication Year: 1998, Page(s):1873 - 1876 vol.3
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (244 KB)

    Some part-of-speech tagging errors are very damaging to the ability to further process the text. For systems that use part-of-speech tagging as a prelude to parsing and knowledge extraction, it is imperative to have the cleanest possible tagging. A state-of-the-art rule-based tagger has an error rate of approximately 39% when annotating main verbs that have not been previously seen. We apply neura... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A temporal adaptive probability neural network for cloud classification from satellite imagery

    Publication Year: 1998, Page(s):1732 - 1737 vol.3
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (684 KB)

    Cloud classification from satellite imagery is an important but very difficult task. Temporal changes are one of the main factors that cause degradation in the classifier performance when a sequence of imagery is to be processed A probability neural network (PNN)-based cloud classification system and its temporal updating scheme is proposed in this paper. This novel approach can track the temporal... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Modeling time dependencies in the mixture of experts

    Publication Year: 1998, Page(s):2324 - 2327 vol.3
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (324 KB)

    The mixture of experts, as it was originally formulated, is a static algorithm in the sense that the output of the network, and parameter updates during training, are completely independent from one time step to the next. This independence creates difficulties when the model is applied to time series prediction. We address this by adding memory to the mixture of experts. A Gaussian assumption on e... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Learning algorithms for reformulated radial basis neural networks

    Publication Year: 1998, Page(s):2230 - 2235 vol.3
    Cited by:  Papers (10)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (444 KB)

    This paper proposes supervised learning algorithms based on gradient descent for training reformulated radial basis function (RBF) neural networks. Such RBF models employ radial basis functions whose form is determined by admissible generator functions. RBF networks with Gaussian radial basis functions are generated by exponential generator functions. A sensitivity analysis provides the basis for ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Clustering noisy data by a principal feature extraction unsupervised neural network

    Publication Year: 1998, Page(s):2361 - 2366 vol.3
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (444 KB)

    Principal feature classification is based on a sequential procedure for finding the principal features from an assigned data set. This paper presents an unsupervised neural network which is able to find principal features, based on neural units sensitive to density of the data space. These units adopt a modified competitive learning law, which utilizes only local information to specialize toward a... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neuro-fuzzy posture estimation for visual vehicle guidance

    Publication Year: 1998, Page(s):2086 - 2091 vol.3
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (708 KB)

    This paper presents a neuro-fuzzy approach to visual guidance of a mobile robot vehicle in local manoeuvres. It is based on the transfer of the skills of an experienced driver to an automatic controller. The resulting controller processes video sensor data to generate corresponding steering and velocity commands in real time. Neither a geometric environment model nor analytic models of the video s... View full abstract»

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