Home     Top: Machine Learning: Neural Networks    [Case-based Learning   Fuzzy Systems   Genetic Algorithms   Neural Networks   Pattern Recognition   Reinforcement Learning   Rule Based Systems   Vision]

Change ordering:   Authority   Hubs (tutorials)   Date   Expected authority       Show titles only
Reverse date order

This directory is created automatically and some papers may be mislabeled. Only document within the CiteSeer database are listed. The directory is intended to provide entry points for browsing the database and is not intended to be authoritative. Papers may not appear in all relevant categories. For example, papers in a sub-category may not appear in higher level categories.

CAM-BRAIN: ATR's ARTIFICIAL BRAIN PROJECT - An Overview - de Garis, Korkin, Gers, Nawa, Hough (2001)   (Correct)
This paper provides an overview of ATR's Artificial Brain (CAM-Brain) Project, which aims to build an artificial brain containing some billion artificial neurons by the year 2001, and to use this brai... / Keywords Artificial Neural Networks Artificial Brain CAM-Brain br evolves cellular automata based neural network modules at electronic speeds

Non-Supervised Sensory-Motor Agents Learning - Wazlawick (2000)   (Correct)
This text initially discusses the distribution of neurons in a neural network with non-supervised learning. A proposal for creation and destruction of neurons based on the activity related to the conc... / algorithms non-supervised neural networks construtivist artificial

Analysis, Visualization and Meta-analysis of Functional Neuroimages.. - Nielsen (1999)   (Correct)
New approaches in this thesis: ffl Comparison of convolution models for fMRI time-series modelling ffl Canonical Ridge Analysis applied to functional neuroimages ffl Some kind of neuroinformatics ... / . . Articial neural networks . br . . Analyzing a neural network the saliency map .

Robust Automatic Speech Recognition With Unreliable Data - Josifovski (1999)   (Correct)
Theoretical and practical issues of some of the problems in robust automatic speech recognition (ASR) and some of the techniques that address them are presented in this report. The problem of the robu... / models HMM and artificial neural networks ANN This is referred to as br imputation with artificial neural networks as posterior estimators .

A Unifying Information-theoretic Framework for Independent Component.. - Lee, Girolami, Bell, Sejnowski (1999)   (Correct)
We show that different theories recently proposed for Independent Component Analysis (ICA) lead to the same iterative learning algorithm for blind separation of mixed independent sources. We review th... / the inputs and outputs of a neural network. This approach is related to br the input and output of a neural network implied that the output

Continuous-based Heuristics for Graph and Tree Isomorphisms, with.. - Pelillo, Siddiqi, Zucker (1999)   (Correct)
We present a new (continuous) quadratic programming approach for graph- and tree-isomorphism problems which is based on an equivalent maximum clique formulation. The approach is centered around a fund... / Durbin and Willshaw the neural network community also became br state-of-the-art sophisticated neural network algorithms which by contrast

Separation of Speech from Interfering Sounds Based on Oscillatory.. - Wang, Brown (1999)   (Correct)
A multi-stage neural model is proposed for an auditory scene analysis task -- segregating speech from interfering sound sources. The core of the model is a twolayer oscillator network that performs st... / of investigators to propose neural network models of ASA. Perhaps the br the first of these was the neural network model described by von der

Fast and Robust Fixed-Point Algorithms for Independent Component.. - Hyvärinen (1999)   (Correct)
Independent component analysis (ICA) is a statistical method for transforming an observed multidimensional random vector into components that are statistically as independent from each other as possib... / To appear in IEEE Trans. on Neural Networks January Abstract br A central problem in neural network research as well as in

Using Neural Networks for Descriptive Statistical Analysis of.. - Tirri (1999)   (Correct)
In this paper we discuss the methodological issues of using a class of neural networks called Mixture Density Networks (MDN) for discriminant analysis. MDN models have the advantage of having a rigoro... / IL USA March Using Neural Networks for Descriptive Statistical br issues of using a class of neural networks called Mixture Density

Efficient exploration for optimizing immediate reward - Schuurmans, Greenwald (1999)   (Correct)
We consider the problem of learning an effective behavior strategy from reward. Although much studied, the issue of how to use prior knowledge to scale optimal behavior learning up to real-world probl... / of machine learning and neural network research. Although a lot of br Subramanian Parr neural networks for learning one shot

Segmentation of Monochrome and Color Textures Using Moving Average.. - Eom (1999)   (Correct)
The segmentation of textures using features extracted with 2-D moving average (MA) modeling approach is presented in this paper. The 2-D MA model represents a texture as an output of a 2-D finite impu... / segmentation is considered. A neural network is used for supervised br windows are classified with a neural network for supervised segmentation

A Mean Field Learning Algorithm For Unsupervised Neural Networks - Lawrence Saul, Michael Jordan (1999)   (Correct)
We introduce a learning algorithm for unsupervised neural networks based on ideas from statistical mechanics. The algorithm is derived from a mean field approximation for large, layered sigmoid beli... / Algorithm For Unsupervised Neural Networks Lawrence Saul At t Labs br algorithm for unsupervised neural networks based on ideas from

Exploring Protein Sequence Space Using Knowledge Based Potentials - Babajide, Farber, Hofacker, Inman.. (1999)   (Correct)
Knowledge-Based potentials can be used to decide whether an amino acid sequence is likely to fold into a prescribed native protein structure. We use this idea to survey the sequencestructure relations... / PROSA pair potential and a neural network based potential are used to br the C backbone. The Neural Network NN Potential includes

SARDSRN: A Neural Network Shift-Reduce Parser - Mayberry, III, Miikkulainen (1999)   (Correct)
Simple Recurrent Networks (SRNs) have been widely used in natural language tasks. SARDSRN extends the SRN by explicitly representing the input sequence in a SARDNET self-organizing map. The distribute... / SARDSRN A Neural Network Shift-Reduce Parser br The subsymbolic approach i.e. neural networks with distributed

Adaptive Optics: Neural Network Wavefront Sensing, Reconstruction and .. - Patrick Mcguire (1999)   (Correct)
We introduce adaptive optics as a technique to improve images taken by ground-based telescopes through a turbulent blurring atmosphere. Adaptive optics rapidly senses the wavefront distortion refere... / Adaptive Optics Neural Network Wavefront Sensing br we summarize the application of neural networks in adaptive optics. First

The Self-Organizing Map in Industry Analysis - Simula, Vasara, Vesanto, Helminen (1999)   (Correct)
The Self-Organizing Map (SOM) is a powerful neural network method for the analysis and visualization of high-dimensional data. It maps nonlinear statistical relationships between high-dimensional meas... / Map SOM is a powerful neural network method for the analysis and br is one of the most popular neural network models. The SOM implements a

Exploiting Domain-Specific Properties: Compiling Parallel Dynamic.. - Prechelt (1999)   (Correct)
Domain-specific constraints can be exploited to implement compiler optimizations that are not otherwise feasible: Compilers for neural network learning algorithms can achieve near-optimal co-locality ... / Compiling Parallel Dynamic Neural Network Algorithms into Efficient br feasible Compilers for neural network learning algorithms can

Telephone Speech Recognition using Neural Networks and Hidden Markov.. - Yuk, Flanagan (1999)   (Correct)
The performance of well-trained speech recognizers using high quality full bandwidth speech data is usually degraded when used in real world environments. In particular, telephone speech recognition i... / Speech Recognition using Neural Networks and Hidden Markov Models br . Maximum Likelihood Neural Network Neural networks are usually trained

Existence and Learning of Oscillations in Recurrent Neural Networks - Townley, Ilchmann, Weiß, Mcclements, .. (1999)   (Correct)
In this paper we study a particular class of n-node recurrent neural networks (RNNs). In the 3-node case we use monotone dynamical systems theory to show, for a well-defined set of parameters, that, g... / of Oscillations in Recurrent Neural Networks S. Townley z-A. br class of n-node recurrent neural networks RNNs In the -node case

Efficient Value Function Approximation Using Regression Trees - Wang, Dietterich (1999)   (Correct)
Value function approximation is a problem central to reinforcement learning. Many applications of reinforcement learning have relied on neural network function approximators, which are very slow to tr... / learning have relied on neural network function approximators which br method performs as well as the neural network method employed in that work

Towards Unrestricted Lip Reading - Meier, Stiefelhagen, Yang, Waibel (1999)   (Correct)
Lip reading provides useful information in speech perception and language understanding, especially when the auditory speech is degraded. However, many current automatic lip reading systems impose som... / Multiple State-Time Delayed Neural Network MS-TDNN system. We have br Another approach is to use a neural network to compute the combination

Adaptive multichannel marginal L-filters - Kotropoulos, Pitas (1999)   (Correct)
Three adaptive multichannel L-filters based on marginal data ordering are proposed. They rely on well-known algorithms for the iterative minimization of the mean square error (MSE), namely, the least ... / white noise. A self-organizing neural network is used to detect the pixels br needed for the self-organizing neural network to converge. In this paper

Fusion Via a Linear Combination of Scores - Vogt, Cottrell (1999)   (Correct)
We present a thorough analysis of the capabilities of the linear combination model for fusion of information retrieval systems. We first present empirical and analytical justification for the hypoth... / linear combination fusion neural networks routing performance br . Introduction In the past neural network models which have been

Learning Decision Trees for Loss Minimization in Multi-Class Problems - Margineantu, Dietterich (1999)   (Correct)
Many machine learning applications require classifiers that minimize an asymmetric loss function rather than the raw misclassification rate. We study methods for modifying C4.5 to incorporate arbitrar... / costs. For neural network algorithms Haykin we br Haykin S. Neural Networks. A Comprehensive Foundation.

Discrimination Of Cylinders With Different Wall Thicknesses Using.. - Andersen, Au, Larsen, Hansen (1999)   (Correct)
This paper describes a method integrating neural networks into a system for recognizing underwater objects. The system is based on a combination of simulated dolphin sonar signals, simulated auditory ... / Wall Thicknesses Using Neural Networks And Simulated Dolphin Sonar br describes a method integrating neural networks into a system for recognizing

Training Reinforcement Neurocontrollers Using The Polytope Algorithm - Likas, Lagaris (1999)   (Correct)
A new training algorithm is presented for delayed reinforcement learning problems that does not assume the existence of a critic model and employs the polytope optimization algorithm to adjust the wei... / model is usually a feedforward neural network trained using the method of br an Inverted Pendulum Using Neural Networks IEEE Control Systems

Properties of Learning of a Fuzzy ART Variant - Georgiopoulos, Dagher, al. (1999)   (Correct)
This paper discusses one variation of the Fuzzy ART algorithm, referred to as Fuzzy ART Variant. The Fuzzy ART Variant is a Fuzzy ART algorithm, with a very large choice parameter value. Based on the ... / choice parameter. Keywords-Neural Network Unsupervised Learning br Architecture The Fuzzy ART neural network architecture is shown in

Evolution and Analysis of Dynamical Neural Networks for Agents.. - John Gallagher (1999)   (Correct)
The use of evolutionary approaches to create dynamical "nervous systems" for autonomous agents is becoming increasingly widespread. In previous work, we have successfully applied this approach to chem... / and Analysis of Dynamical Neural Networks for Agents Integrating br and analyzing dynamical neural networks for visually-guided walking.

A Comparative Study of Neural Network Based Feature Extraction.. - Lerner, Guterman, Aladjem, Dinstein (1999)   (Correct)
The projection maps and derived classification accuracies of a neural network (NN) implementation of Sammon's mapping, an auto-associative NN (AANN) and a multilayer perceptron (MLP) feature extractor... / A Comparative Study of Neural Network Based Feature Extraction br classification accuracies of a neural network NN implementation of

An Instructable, Adaptive Interface for Discovering and Monitoring.. - Jude Shavlik (1999)   (Correct)
We are creating a customizable, intelligent interface to the World-Wide Web that assists a user in locating specific, current, and relevant information. The Wisconsin Adaptive Web Assistant (Wawa) is ... / these instructions into neural networks which means that the br .Wawa uses these neural networks to guide its autonomous

Regularizing AdaBoost - Rätsch, Onoda, Müller (1999)   (Correct)
Boosting methods maximize a hard classification margin and are known as powerful techniques that do not exhibit overfitting for low noise cases. Also for noisy data boosting will try to enforce a hard... / be an outlier. In general e.g. neural network learning strategies this br C. M. Bishop. Neural Networks for Pattern Recognition.

On the Momentum Term in Gradient Descent Learning Algorithms - Ning Qian (1999)   (Correct)
A momentum term is usually included in the simulations of connectionist learning algorithms. Although it is well known that such a term greatly improves the speed of learning, there have been few rigo... / To appear in Neural Networks On the Momentum Term in br Introduction Connectionist neural network models have been successfully

Feature Extraction and Learning Vector Quantization for Data.. - Goller, Gori (1999)   (Correct)
During the last years, folding architecture networks and the closely related concept of recursive neural networks have been developed for solving supervised learning tasks on data structures. In this ... / related concept of recursive neural networks have been developed for br pattern recognition or neural network approaches. During the last

Feature Reduction for Neural Network Based Text Categorization - Savio Lam (1999)   (Correct)
In a text categorization model using an artificial neural network as the text classifier, scalability is poor if the neural network is trained using the raw feature space since textural data has a ver... / Feature Reduction for Neural Network Based Text Categorization br model using an artificial neural network as the text classifier

Co-operative Evolution of a Neural Classifier and Feature Subset - Hallinan, Jackway (1999)   (Correct)
This paper describes a novel feature selection algorithm which utilizes a genetic algorithm to select a feature subset in conjunction with the weights for a three-layer feedforward network classifie... / input features and a simpler neural network architecture. These results br indicate that tailoring a neural network classifier to a specific

Fuzzy Inputs and Missing Data in Similarity-Based Heterogeneous.. - Belanche, Valdés (1999)   (Correct)
Fuzzy heterogeneous networks are recently introduced feed-forward neural network models composed of neurons of a general class whose inputs and weights are mixtures of continuous variables (crisp and/... / Similarity-Based Heterogeneous Neural Networks Llu'is A. Belanche Julio J. br introduced feed-forward neural network models composed of neurons of

Using Attribute Grammars for the Genetic Selection of Backpropagation .. - Roger Browse (1999)   (Correct)
Determining exactly which neural network architecture, with which parameters, will provide the best solution to a classification task is often based upon the intuitions and experience of the implement... / Determining exactly which neural network architecture with which br of the implementers of neural network solutions. The research

Facial Feature Extraction using Deformable Graphs and Statistical.. - Ahlberg (1999)   (Correct)
In model-based coding of image sequences containing human faces, e.g., videophone sequences, the detection and location of the face as well as the extraction of facial features from the images are cru... / connectionist neural network methods or statistical

Mean field methods for classification with Gaussian processes - Opper, Winther (1999)   (Correct)
We discuss the application of TAP mean field methods known from the Statistical Mechanics of disordered systems to Bayesian classification models with Gaussian processes. In contrast to previous appro... / as a limit of a two-layered neural network with infinitely many hidden br Derivations Of The Tap Mft For Neural Networks Was The Fact That Special

The Thisl Broadcast News Retrieval System - Dave Abberley (1999)   (Correct)
This paper described the THISL spoken document retrieval system for British and North American Broadcast News. The system is based on the ABBOT large vocabulary speech recognizer, using a recurrent ne... / estimates produced by the neural network acoustic model by pruning all

Switching Controllers Based on Neural Network Estimates of Stability.. - Ferreira, Krogh (1999)   (Correct)
This paper presents new results on switching control using neural networks. Given a set of candidate controllers, a pair of neural networks is trained to identify the stability region and estimate t... / Switching Controllers Based on Neural Network Estimates of Stability br on switching control using neural networks. Given a set of candidate

A Multi-agent based Evolutionary Artificial Neural Network for.. - Wang, Mckenzie (1999)   (Correct)
This paper presents a multi-agent based evolutionary artificial neural network (ANN) for general navigation. While vision is a single input channel to the ANN, only the information about the availabil... / based Evolutionary Artificial Neural Network for General Navigation in br based evolutionary artificial neural network ANN for general navigation.

Automatic Discrimination Among Languages Based on Prosody Alone - Fred Cummins, Felix Gers, Jürgen.. (1999)   (Correct)
The development of methods for the automatic identification of languages is motivated both by speech-based applications intended for use in a multi-lingual environment, and by theoretical questions of... / is done using a novel neural network which can successfully attend br series to a novel recurrent neural network. The network employed has

Learning Nonlinear Dynamical Systems using an EM Algorithm - Ghahramani, Roweis (1999)   (Correct)
The Expectation--Maximization (EM) algorithm is an iterative procedure for maximum likelihood parameter estimation from data sets with missing or hidden variables [2]. It has been applied to system id... / if they are represented by neural network regressors a single full M

Nonlinear Autoassociation is not Equivalent to PCA - Nathalie Japkowicz (1999)   (Correct)
A common misperception within the Neural Network community is that even with nonlinearities in their hidden layer, autoassociators trained with Backpropagation are equivalent to linear methods such as... / common misperception within the Neural Network community is that even with br A typical belief in the Neural Network community today is that

Active Learning in Self-Organizing Maps - Hasenjager, Ritter, Obermayer (1999)   (Correct)
ore than a factor of two compared to a random selection strategy. This makes active data selection a viable alternative when the cost of actually measuring dissimilarities between data objects becomes... / Figure . Left Sketch of a neural network autoencoder with three hidden br the sketch of a feedforward neural network architecture called

Relaxed Simulated Tempering for VLSI Floorplan Designs - Cong, Kong, Xu, Liang, Liu, Wong (1999)   (Correct)
In the past two decades, the simulated annealing technique has been considered as a powerful approach to handle many NP-hard optimization problems in VLSI designs. Recently, a new Monte Carlo and opti... / traveling salsman problem and neural network training. In this paper we

No Free Lunch for Early Stopping - Cataltepe, Abu-Mostafa, Magdon-Ismail (1999)   (Correct)
We show that, with a uniform prior on models having the same training error, early stopping at some fixed training error above the training error minimum results in an increase in the expected general... / models one hidden layer neural networks with a linear output and br results for general linear and neural network models. We also compare weight

On Mapping Decision Trees and Neural Networks - Rudy Setiono (1999)   (Correct)
There exist several methods for transforming decision trees to neural networks. These methods typically construct the networks by directly mapping decision nodes or rules to the neural units. As a res... / On Mapping Decision Trees and Neural Networks Rudy Setiono and Wee Kheng br transforming decision trees to neural networks. These methods typically

On Condition Monitoring Of Exhaust Valves In Marine Diesel Engines - Fog, Hansen, Larsen, Hansen, Madsen, .. (1999)   (Correct)
Experimental investigations of the possibilities to non-intrusively characterize exhaust valve conditions in large marine diesel engines, were carried out on a four cylinder, 500 mm bore 2-stroke ma... / The complexity of the neural networks have further been optimized br optimal classifier using a neural network which outputs estimates of the

Search in a Small World - Walsh (1999)   (Correct)
In a graph with a "small world" topology, nodes are highly clustered yet the path length between them is small. Such a topology can make search problems very difficult since local decisions quickly pr... / a biological graph the neural network of the nematode worm C.

An Fft-Based Algorithm For Multichannel Blind Deconvolution - Joho, Mathis, Moschytz (1999)   (Correct)
A new update equation for the general multichannel blind deconvolution (MCBD) of a convolved mixture of source signals is derived. It is based on the update equation for blind source separation (BSS),... / be understood as a single-layer neural network whose training algorithm and

An Agent Infrastructure for Knowledge Discovery and Event Detection - Martin, Unruh, Urban (1999)   (Correct)
Data mining and data analysis is often a sub-component of a larger knowledge discovery process within an organization. Although toolkits for data mining generally provide some form of support for know... / For example one might apply a neural network to credit card transaction

Monitoring Piecewise Continuous Behaviors by Refining Trackers and.. - Rinner (1999)   (Correct)
We present a model-based monitoring method for dynamic systems that exhibit both discrete and continuous behaviors. MIMIC (Dvorak & Kuipers 1991) uses qualitative and semi-quantitative models to monit... / are generated by MSQUID a neural network-based estimator for monotonic

Detecting Hunts in Wildlife Videos - Haering, Qian, Sezan (1999)   (Correct)
We propose a three-level algorithm to detect animal hunt events in wildlife documentaries. The first level extracts texture, color and motion features, and detects motion blobs. The mid-level employs ... / blobs. The mid-level employs a neural network to verify the relevance of br as moving object regions by a neural network. The network uses the color

Multi-modal Stereognosis - Goncalves, Grupen, Oliveira (1999)   (Correct)
Introduction In this work, vision and touch (artificial) senses are integrated in a cooperative active system. Multi-modal sensory information acquired on-line is used by a robotic agent to perform r... / each arm. A back-propagation neural network implements the associative

A Low Power VLSI Arrhythmia Classifier - Leong, Jabri (1999)   (Correct)
The design, implementation and operation of a low power multilayer perceptron chip (Kakadu) in the framework of a cardiac arrhythmia classification system is presented in this paper. This system, call... / Ieee Transactions On Neural Networks Vol. Xx No. Y Month br Draft Ieee Transactions On Neural Networks Vol. Xx No. Y Month

Language Identification From Prosody Without Explicit Features - Fred Cummins, Felix Gers, Jürgen.. (1999)   (Correct)
Most current language identification (LID) systems make little or no use of prosodic information, despite the importance of prosody in LID by humans. The greatest obstacle has been that of finding an ... / We apply a novel recurrent neural network model to the task of pairwise br Identification Recurrent Neural Networks Prosody . Prosody And

A New Approach to Kanerva's Sparse Distributed Memory - Hely, Willshaw, Hayes (1999)   (Correct)
The Sparse Distributed Memory (SDM)[1] was originally developed to tackle the problem of storing large binary data patterns. The model succeeded well in storing random input data. However its efficien... / Ieee Transactions On Neural Networks Vol. Xx No. Y Month br the memory as a single-layer neural network.See also

The Early Restart Algorithm - Magdon-Ismail, Atiya (1999)   (Correct)
Consider an algorithm whose time till convergence is unknown (because of some random element in the algorithm, such as for example a random initial weight choice for neural network training). Consider... / initial weight choice for neural network training Consider the br to training algorithms in the neural network field. Analysis of the

Building An Artificial Brain Using An FPGA Based "CAM-Brain Machine" - de Garis, GERS, KORKIN, AGAH, NAWA (1999)   (Correct)
This paper reports on recent progress made in ATR's attempt to build a 10,000 evolved neural net module artificial brain to control the behaviors of a life sized robot kitten. 1. Introduction This pap... / upon which to grow and evolve neural network circuits with user defined br was how to evolve these neural networks quickly enough for Brain

Predicting the Speed of Beer Fermentation in Laboratory and.. - Rousu, Elomaa, Aarts (1999)   (Correct)
Characteristic of the beer production process is the uncertainty caused by the complex biological raw materials and the yeast, a living organism. This uncertainty is exempliøed by the fact that pred... / a non-trivial task. We employ neural network and decision tree learning to br laboratory-scale experiments a neural network that employs characteristics

Design of a Genetic-Fuzzy System for Planning Crab Gaits of a.. - Pratihar, Deb, Ghosh (1999)   (Correct)
This paper describes a genetic-fuzzy system in which a genetic algorithm (GA) is used to improve the performance of a fuzzy logic controller (FLC). The proposed algorithm is tested on a number of gait... / gait generation problem using neural network NN but in their approach br As training is required in neural network its computational complexity

INSS : an hybrid system for constructive machine learning - Fernando Osorio (1999)   (Correct)
In this paper we present the INSS system, a new hybrid approach based upon the principles of KBANN networks. It represents an important improvement in comparison with its predecessor because the learn... / algorithm Artificial Neural Networks ANN rule insertion and br methods Artificial Neural Networks -ANN Such a justification

A Note on a Capacity Result for the Hopfield Neural Network - Pelillo (1999)   (Correct)
In this note we prove, by using straightforward arguments, a stronger version of a well-known capacity result proved by Abu-Mostafa and St. Jacques. Our results states that no Hopfield memory can have... / Result for the Hopfield Neural Network Marcello Pelillo br of the discrete Hopfield neural network. Theorem Let p

Vector-Based Natural Language Call Routing - Chu-Carroll, Carpenter (1999)   (Correct)
This paper describes a domain independent, automatically trained natural language call router for directing incoming calls in a call center. Our call router directs customer calls based on their respo... / weighted fragments through a neural network classifier Wright Gorin and

Evolving an Optimal De/Convolution Function for the Neural Net.. - de Garis, Nawa, Hough, Korkin (1999)   (Correct)
This paper reports on efforts to evolve an optimum de/convolution function to be used to convert analog to binary signals (spike trains) and vice versa for the binary input/output signals of the neu... / of cellular automata based neural network circuits or modules at br cellular automata CA based neural network circuits modules at

A Framework for Programming Embedded Systems: Initial Design and.. - Thrun (1998)   (Correct)
This paper describes CES, a proto-type of a new programming language for robots and other embedded systems, equipped with sensors and actuators. CES contains two new ideas, currently not found in othe... / system trained an artificial neural network to map camera images to br own work we recently employed neural networks for sensor interpretation and

Incorporating Information From Syllable-length Time Scales Into.. - Wu (1998)   (Correct)
Incorporating the concept of the syllable into speech recognition may improve recognition accuracy through the integration of information over syllable-length time spans. Evidence from psychoacoustics... / . . Probability Estimation Neural Network br from an onset- detection neural network at frames covered by

Optimization and global minimization methods suitable for neural.. - Duch, Korczak (1998)   (Correct)
Neural networks are usually trained using local, gradient-based procedures. Such methods frequently find suboptimal solutions being trapped in local minima. Optimization of neural structures and globa... / methods suitable for neural networks Wl odzisl aw Duch and br Illkirch France Abstract Neural networks are usually trained using

Minimal Simulations For Evolutionary Robotics - Jakobi (1998)   (Correct)
this paper, the line is drawn between controller and environment. 3.1.1 Drawing the line between controller and environment unknown Minimal Simulations For Evolutionary Robotics Nick Jakobi Submitted... / discusses the best types of neural network encoding scheme genetic br . . . Neural networks for evolutionary robotics .

A Brief History of Connectionism - Medler (1998)   (Correct)
Connectionist research is firmly established within the scientific community, especially within the multi-disciplinary field of cognitive science. This diversity, however, has created an environment w... / concludes by suggesting that neural network research-at least in br and Olson have stated The neural network revolution has happened. We

Modular Connectionist Architectures and the Learning of.. - Bale (1998)   (Correct)
Modular connectionist systems comprise autonomous, communicating modules, achieving a behaviour more complex than that of a single neural network. The component modules, possibly of different topologi... / complex than that of a single neural network. The component modules br . Modular Neural

Natural Language Grammatical Inference with Recurrent Neural Networks - Lawrence, Giles, Fong (1998)   (Correct)
This paper examines the inductive inference of a complex grammar with neural networks -- specifically, the task considered is that of training a network to classify natural language sentences as gramm... / Inference with Recurrent Neural Networks Steve Lawrence C. Lee br of a complex grammar with neural networks -specifically the task

Self-Organizing Maps And Software Reuse - Merkl (1998)   (Correct)
Software reuse is the process of building new systems from existing components instead of developing these systems from scratch. For a long time now software reuse is repeatedly acknowledged for playi... / map i.e. an unsupervised neural network model to determine the mutual br of selforganizing maps the neural network model we rely on for software

Fuzzy Finite-state Automata Can Be Deterministically Encoded into.. - Omlin, Thornber, Giles (1998)   (Correct)
There has been an increased interest in combining fuzzy systems with neural networks because fuzzy neural systems merge the advantages of both paradigms. On the one hand, parameters in fuzzy systems h... / Encoded Into Recurrent Neural Networks Christian W. Omlin a br in combining fuzzy systems with neural networks because fuzzy neural systems

How to Build VLSI-Efficient Neural Chips - Beiu (1998)   (Correct)
This paper presents several upper and lower bounds for the number-of-bits required for solving a classification problem, as well as ways in which these bounds can be used to efficiently build neural n... / be used to efficiently build neural network chips. The focus will be on br aspects pertaining to neural networks i size complexity and

Map Learning and High-Speed Navigation in RHINO - Thrun, Bücken, Burgard, Fox.. (1998)   (Correct)
This chapter surveys basic methods for learning maps and high speed autonomous navigation for indoor mobile robots. The methods have been developed in our lab over the past few years, and most of them... / algorithms such as artificial neural networks interpret the robot's br and the cameras. Artificial neural networks are used to interpret sonar

Constructive Neural Network Learning Algorithms for Pattern.. - Parekh, Yang, Honavar (1998)   (Correct)
Constructive learning algorithms offer an attractive approach for the incremental construction of near-minimal neural network architectures for pattern classification. They help overcome the need for ... / Constructive Neural Network Learning Algorithms for br construction of near-minimal neural network architectures for pattern

Relational Learning Techniques for Natural Language Information.. - Califf (1998)   (Correct)
The recent growth of online information available in the form of natural language documents creates a greater need for computing systems with the ability to process those documents to simplify access ... / significant research applying neural-network methods to language processing

The Role of Afferent Excitatory and Lateral Inhibitory Synaptic.. - Kalarickal, Marshall (1998)   (Correct)
Previous models of visual cortical ocular dominance (OD) plasticity (e.g., Clothiaux et al., 1991; Miller et al., 1989) are based on afferent excitatory synaptic plasticity alone; these models do not ... / plasticity self-organization neural networks EXIN excitatory br We have formulated and tested a neural network model that exhibits OD changes

Hierarchical Bayesian-Kalman Models For Regularisation And ARD In.. - de Freitas, Niranjan, Gee (1998)   (Correct)
In this paper, we show that a hierarchical Bayesian modelling approach to sequential learning leads to many interesting attributes such as regularisation and automatic relevance determination. We iden... / Sequential training of neural networks is important in applications br Sutton's ideas to nonlinear neural networks and relate them to other

Empirical Risk Approximation: An Induction Principle for Unsupervised .. - Buhmann (1998)   (Correct)
Unsupervised learning algorithms are designed to extract structure from data without reference to explicit teacher information. The quality of the infered structure is determined by a quality function... / al. and various other neural network models Poggio and Girosi br Cambridge UK under the Neural Networks and Machine Learning

Automatic Text Detection and Tracking in Digital Video - Li, Doermann, Kia (1998)   (Correct)
Text which either appears in a scene or is graphically added to video can provide an important supplemental source of index information as well as clues for decoding the video's structure and for clas... / Indexing Digital Libraries Neural Network Wavelet The support of br We use a hybrid wavelet neural network segmenter Figure to

Feature Weighting for Lazy Learning Algorithms - Aha (1998)   (Correct)
Learning algorithms differ in the degree to which they process their inputs prior to their use in performance tasks. Many algorithms eagerly compile input samples and use only the compilations to m... / a rule set decision tree or neural network and use it to process br Expert Systems and Neural Networks pages - . Unpublished.

SemQuery: Semantic Clustering and Querying on Heterogeneous Features.. - Sheikholeslami, Chang, Zhang (1998)   (Correct)
The effectiveness of the content-based image retrieval can be enhanced using the heterogeneous features embedded in the images. However, since the features in texture, color, and shape are generated u... / We also design a multi-layer neural network model to merge the results of br features. The input to the neural network is the set of image features

A Study of the Use and Evaluation of Confidence Measures in Automatic .. - Williams (1998)   (Correct)
Confidence measures have been found to be useful for a number tasks within the field of Automatic Speech Recognition (ASR). For example, the use of confidence measures has been reported in the utteran... / Hidden Markov Model Artificial Neural Network HMM ANN systems are well br Hidden Markov Model Artificial Neural Network HMM ANN based

Equivalence in Knowledge Representation: Automata, Recurrent Neural.. - Omlin, Giles, Thornber (1998)   (Correct)
Neuro-fuzzy systems - the combination of artificial neural networks with fuzzy logic - are becoming increasingly popular. However, neuro-fuzzy systems need to be extended for applications which requir... / Automata Recurrent Neural Networks and Dynamical Fuzzy Systems br -the combination of artificial neural networks with fuzzy logic -are

Learning Maps for Indoor Mobile Robot Navigation - Thrun, Bücken (1998)   (Correct)
Autonomous robots must be able to learn and maintain models of their environments. Research on mobile robot navigation has produced two major paradigms for mapping indoor environments: grid-based and ... / exploration mobile robots neural networks occupancy grids path br are learned using artificial neural networks and Bayesian integration.

Generalization - Wah (1998)   (Correct)
In this paper, we define the generalization problem, summarize various approaches in generalization, identify the credit assignment problem, and present the problem and some solutions in measuring gen... / systems and artificial neural networks ANNs In contrast to br truck backer and neural-network design The

Networks For Speech Enhancement - Wan, Nelson (1998)   (Correct)
Introduction 1.1. Background Speech enhancement is motivated by the need to improve the performance of voice communications systems in noisy conditions. Applications range from frontends for speech ... / Handbook of Neural Networks for Speech Processing First br recent years several distinct neural network based frameworks have emerged

A Tutorial on Support Vector Machines for Pattern Recognition - Burges (1998)   (Correct)
The tutorial starts with an overview of the concepts of VC dimension and structural risk minimization. We then describe linear Support Vector Machines (SVMs) for separable and non-separable data, wo... / leave the best performing neural networks severely handicapped and br machine.Thus for example a neural network with fixed architecture with

Sparse coding with an overcomplete basis set: A strategy employed by.. - Olshausen, Field (1998)   (Correct)
The spatial receptive fields of simple cells in mammalian striate cortex have been reasonably well described physiologically and can be characterized as being localized, oriented, and bandpass, compa... / by a strictly feedforward neural network in which case the functions br as shown in Figure b. A neural network implementation of this

Bayesian deviance, the effective number of parameters, and the.. - Spiegelhalter, Best, Carlin (1998)   (Correct)
We consider the problem of comparing complex hierarchical models in which the number of parameters is not clearly defined. We follow Dempster in examining the posterior distribution of the log-likelih... / particular reference to the neural network literature. The contributions br this problem appear in the neural network literature Moody

Making Use of Population Information in Evolutionary Artificial.. - Yao, Liu (1998)   (Correct)
This paper is concerned with the simultaneous evolution of artificial neural network (ANN) architectures and weights. The current practice in evolving ANNs is to choose the best ANN in the last genera... / in Evolutionary Artificial Neural Networks Xin Yao SMIEEE and Yong br evolution of artificial neural network ANN architectures and

Bayes Optimal Instance-Based Learning - Kontkanen, Myllymäki, al. (1998)   (Correct)
In this paper we present a probabilistic formalization of the instance-based learning approach. In our Bayesian framework, moving from the construction of an explicit hypothesis to a data-driven ins... / include the set of feedforward neural network models the set of Bayesian br include the set of feedforward neural network models the set of Bayesian

Learning to Perceive the World as Articulated: An Approach for.. - Tani, Nolfi (1998)   (Correct)
This paper describes how agents can learn an internal model of the world structurally by focusing on the problem of behavior-based articulation. We develop an on-line learning scheme -- the so-called ... / version is in press. in Neural Networks Special Issue . br types of approach. One is the neural network learning approach. Krose and

BISMARC: A Biologically Inspired System for Map-based Autonomous.. - Huntsberger, Rose (1998)   (Correct)
As the complexity of the missions to planetary surfaces increases, so too does the need for autonomous rover systems. This need is complicated by the power, mass and computer storage restrictions on s... / hippocampal maps wavelets neural networks . INTRODUCTION The br There are also a number of neural network approaches that have been used

Natural Gradient Works Efficiently in Learning - Amari (1998)   (Correct)
When a parameter space has a certain underlying structure, the ordinary gradient of a function does not represent its steepest direction but the natural gradient does. Information geometry is used for... / that parameter spaces of neural networks have the Riemannian br by a processor like a neural network which has a set of adjustable

The Link Between Brain Learning, Attention, And Consciousness - Grossberg (1998)   (Correct)
The processes whereby our brains continue to learn about a changing world in a stable fashion throughout life are proposed to lead to conscious experiences. These processes include the learning of top... / attention adaptive resonance neural network procedural memory br always been interpretable as a neural network. These neural networks have

Optimizing Neural Networks for Time Series Prediction - De Falco, Cioppa, Iazzetta, Natale.. (1998)   (Correct)
In this paper we investigate the effective design of an appropriate neural network model for time series prediction based on an evolutionary approach. In particular, the Breeder Genetic Algorithms are... / Optimizing Neural Networks for Time Series Prediction br Series Prediction Artificial Neural Networks Evolutionary Algorithms

Finite State Machines and Recurrent Neural Networks - Automata and.. - Tino, Horne, Giles (1998)   (Correct)
We present two approaches to the analysis of the relationship between a recurrent neural network (RNN) and the finite state machine M the network is able to exactly mimic. First, the network is treate... / State Machines and Recurrent Neural Networks -Automata and Dynamical

Soft Margins for AdaBoost - Rätsch, Onoda, Müller (1998)   (Correct)
Recently ensemble methods like AdaBoost were successfully applied to character recognition tasks, seemingly defying the problems of overfitting. This paper shows that although AdaBoost rarely overfits... / An ensemble is a collection of neural networks or other types of classifiers br t T . Train neural network with respect to the weighted

Plasticity of directional place fields in a model of rodent CA3 - Brunel, Trullier (1998)   (Correct)
We propose a computational model of the CA3 region of the rat hippocampus that is able to reproduce the available experimental data concerning the dependence of directional selectivity of the place ce... / France Running title Neural network model of rodent CA Number of br brunel lps.ens.fr Keywords neural network model place cells

Constructive Theory Refinement in Knowledge Based Neural Networks - Parekh, Honavar (1998)   (Correct)
Knowledge based artificial neural networks offer an approach for connectionist theory refinement. We present an algorithm for refining and extending the domain theory incorporated in a knowledge based... / Refinement in Knowledge Based Neural Networks Rajesh Parekh Vasant br Knowledge based artificial neural networks offer an approach for

Task Decomposition and Module Combination Based on Class Relations: A .. - Lu, Ito (1998)   (Correct)
In this paper, we propose a new method for decomposing pattern classification problems based on the class relations among training data. By using this method, we can divide a K- class classification ... / on Class Relations A Modular Neural Network for Pattern Classification br in using artificial neural networks for solving large-scale

Feature Subset Selection Using A Genetic Algorithm - Yang, Honavar (1998)   (Correct)
Practical pattern classification and knowledge discovery problems require selection of a subset of attributes or features (from a much larger set) to represent the patterns to be classified. This pape... / in the automated design of neural networks for pattern classification br tree induction algorithm or a neural network learning algorithm The

Issues in Bayesian Analysis of Neural Network Models - Müller, Insua (1998)   (Correct)
This paper discusses these issues exploring the potentiality of Bayesian ideas in the analysis of NN models. Buntine and Weigend (1991) and MacKay (1992) have provided frameworks for their Bayesian an... / Issues in Bayesian Analysis of Neural Network Models Peter M uller br in Bayesian analysis of Neural Network Models. We study

Sparse Code Shrinkage: Denoising of Nongaussian Data by Maximum.. - Hyvärinen (1998)   (Correct)
Sparse coding is a method for finding a representation of data in which each of the components of the representation is only rarely significantly active. Such a representation is closely related to re... / is a method for nding a neural network representation of br vector that is input to a neural network and by s s s

Diagnostic Screening of Digital Mammograms Using Wavelets and Neural.. - Kalman, Kwasny, Reinus (1998)   (Correct)
As the primary tool for detecting breast carcinoma, mammography provides visual images from which a trained radiologist can identify suspicious areas that suggest the presence of cancer. We describ... / Mammograms Using Wavelets and Neural Networks to Extract Structure BARRY br using wavelets and neural networks. To illustrate its utility

On Natural Life's Tricks to Survive and Evolve - Schwefel, Kursawe (1998)   (Correct)
Which are the fundamental principles of life? This is the main question to be addressed if one tries to create artificial life on computers. Though it has been answered only partially, evolutionary al... / EP Society in the IEEE Neural Network Council annual International

On Analog Implementation Of Discrete Neural Networks - Beiu, Moore (1998)   (Correct)
The paper will show that in order to obtain minimum size neural networks (i.e., size-optimal) for implementing any Boolean function, the nonlinear activation function of the neurons has to be the ide... / Implementation Of Discrete Neural Networks V. Beiu A K.r. br order to obtain minimum size neural networks i.e.size-optimal

Efficient and Adaptive Lagrange-Multiplier Methods for Nonlinear.. - Wah, Wang (1998)   (Correct)
Lagrangian methods are popular in solving continuous constrained optimization problems. In this paper, we address three important issues in applying Lagrangian methods to solve optimization problems w... / engineering design neural-network learning computer-aided br in signal processing neural-network learning and benchmark

Forming Neural Networks through Efficient and Adaptive Coevolution - Moriarty, Miikkulainen (1998)   (Correct)
This article demonstrates the advantages of a cooperative, coevolutionary search in difficult control problems. The SANE system coevolves a population of neurons that cooperate to form a functioning n... / Forming Neural Networks through Efficient and br to form a functioning neural network. In this process neurons

Principles of Cortical Processing Applied to and Motivated by.. - Krüger, Pötzsch, Peters (1998)   (Correct)
In this paper we discuss the biological plausibility of the object recognition system described in detail in (Kruger, Peters and v.d. Malsburg, 1996). We claim that this system realizes the following ... / On the other hand the neural network community may loose br E.and Doursat R. Neural networks and the bias variance

Discontinuities in Recurrent Neural Networks - Gavalda, Siegelmann (1998)   (Correct)
This paper studies the computational power of various discontinuous real computational models that are based on the classical analog recurrent neural network (ARNN). This ARNN consists of finite numbe... / Discontinuities in Recurrent Neural Networks Ricard Gavald a Department br the classical analog recurrent neural network ARNN This ARNN consists of

Relaxation Labeling Using Augmented Lagrange-Hopfield Method - Stan Li (1998)   (Correct)
This paper presents a novel relaxation labeling method called Augmented Lagrangian-Hopfield (ALH) method based on the Augmented Lagrangian multipliers and the graded Hopfield neural network. In the AL... / and the graded Hopfield neural network ALH In this method an RL br and the graded Hopfield neural network. In the ALH method RL is

Integrating Reactive and Reflective Reasoning by Generating Rational.. - Bornscheuer (1998)   (Correct)
We propose to model integrated reflective and reactive reasoning by massively parallel nonmonotonic model generation. To this end, a finite representation of models of normal logic programs is given... / with learning are artificial neural networks and wrt the refinement of br given fuzzy descriptions fuzzy neural networks cf. Moreover a

GTM: The Generative Topographic Mapping - Bishop, al. (1998)   (Correct)
Latent variable models represent the probability density of data in a space of several dimensions in terms of a smaller number of latent, or hidden, variables. A familiar example is factor analysis wh... / for example of a feed-forward neural network in which case W would br by EPSRC grant GR K Neural Networks for Visualization of

Optimized Combination, Regularization, and Pruning in Parallel.. - Benediktsson, Larsen, Sveinsson.. (1998)   (Correct)
Optimized combination, regularization, and pruning is proposed for the Parallel Consensual Neural Networks (PCNNs) which is a neural network architecture based on the consensus of a collection of stag... / Pruning in Parallel Consensual Neural Networks J. A. Benediktsson a br for the Parallel Consensual Neural Networks PCNNs which is a neural

Image Thresholding by Indicator Kriging - Oh (1998)   (Correct)
We consider the problem of segmenting a digitized image consisting of two univariate populations. Assume a-priori knowledge allows incomplete assignment of voxels in the image, in the sense that a fra... / Markov random field or neural network based methods edge

Space-Time Modelling Without Distance - Denison, Dellaportas, Mallick (1998)   (Correct)
We present a novel method for analysing space-time data when response data is given at a finite number of locations and the aim is to predict the response at a new location, where only a short run of ... / Any nonparametric model neural network projection pursuit

Independent Component Analysis: A flexible non-linearity and.. - Richard Everson (1998)   (Correct)
Independent Components Analysis finds a linear transformation to variables which are maximally statistically independent. We examine ICA and algorithms for finding the best transformation from the poi... / accomplished by a single layer neural network in which the elements of W

Error Bounds for Functional Approximation and Estimation Using.. - Zeevi, Meir, Maiorov (1998)   (Correct)
In this paper we examine some mathematical aspects of learning unknown mappings with the Mixture of Experts Model (MEM). Specifically, we observe that the MEM is equivalent to a class of neural networ... / MEM is equivalent to a class of neural networks with normalized sigmoidal br the MEM is equivalent to a neural network model in scope while ate the

Evolving and Breeding Robots - Lund, Miglino (1998)   (Correct)
Our experiences with a range of evolutionary robotic experiments have resulted in major changes to our set-up of artificial life experiments and our interpretation of observed phenomena. Initially, we... / work with evolution of neural network agents to real robots. This is br agents e.g. controlled by neural networks to model

An Evaluation of Statistical Approaches to Text Categorization - Yang (1998)   (Correct)
This paper focuses on a comparative evaluation of a wide-range of text categorization methods, including previously published results on the Reuters corpus and new results of additional experiments... / observation kNN LLSF and a neural network method had the best br learning algorithms neural networks and on-line learning

Automatic Early Stopping Using Cross Validation: Quantifying the.. - Lutz Prechelt (1998)   (Correct)
Cross validation can be used to detect when overfitting starts during supervised training of a neural network; training is then stopped before convergence to avoid the overfitting ("early stopping"). ... / Appeared in Neural Networks Automatic Early br supervised training of a neural network training is then stopped

Support Vector Machines for Classification and Regression - Gunn (1998)   (Correct)
this report the term SVM will refer to both classification and regression methods, and the terms Support Vector Classification (SVC) and Support Vector Regression (SVR) will be used for specification.... / of Hadamard Traditional neural network approaches have suffered br employed by conventional neural networks. SRM minimises an upper

Recurrent neural networks with Iterated Function Systems dynamics - Peter Tino, Georg Dorffner (1998)   (Correct)
We suggest a recurrent neural network (RNN) model with a recurrent part corresponding to iterative function systems (IFS) introduced by Barnsley [1] as a fractal image compression mechanism. The key i... / Recurrent neural networks with Iterated Function br We suggest a recurrent neural network RNN model with a recurrent

Neural Network Classification and Prior Class Probabilities - Lawrence, Burns, Back, Tsoi, Giles (1998)   (Correct)
A commonly encountered problem in MLP (multi-layer perceptron) classification problems is related to the prior probabilities of the individual classes - if the number of training examples that corresp... / Verlag pp. - . Neural Network Classi cation and Prior Class br simpler than reformulating the neural network training algorithm for the new

Generalization and exclusive allocation of credit in unsupervised.. - Marshall, Gupta (1998)   (Correct)
A new way of measuring generalization in unsupervised learning is presented. The measure is based on an exclusive allocation, or credit assignment , criterion. In a classifier that satisfies the crite... / in several simple neural network classifiers. The essence of br for benchmarking unsupervised neural network classifier performance have

Multiple-Agent Architectures For The Classification Of Handwritten.. - Louis Vuurpijl (1998)   (Correct)
this paper. The concept of intelligent agents and innovative multi-agent architectures for pattern recognition tasks is introduced for combining and elaborating the classification hypotheses of severa... / in the form of e.g.a neural network a statistical classifier br If trainable statistical or neural-network metaclassifiers are used

Overview of complexity and decidability results for three classes of.. - Blondel, Tsitsiklis (1998)   (Correct)
It has become increasingly apparent this last decade that many problems in systems and control are NP-hard and, in some cases, undecidable. The inherent complexity of some of the most elementary probl... / nonlinearity systems of the neural network type and piecewise-linear br is open. Systems of the neural network type Let us fix a scalar

Integrating Iconic and Structured Matching - Fisher And (1998)   (Correct)
Several investigations [11, 16, 19--21] have recently been undertaken into object recognition based on matching image intensity neighborhoods rather than geometric matching of features extracted fro... / vectors are input into a simple neural network which associates each br Selecting features for neural networks to aid an iconic search

A Statistical Study on On-line Learning - Murata (1998)   (Correct)
In this paper we examine on-line learning with statistical framework. Firstly we study the cases with fixed and annealed learning rate. It can be shown that on-line learning with 1=t annealed learning... / have to note that in artificial neural network learning a problem of local br statistical properties of neural network learning. In Proceedings of

Combining Multiple Views and Temporal Associations for 3-D Object.. - Massad, Mertsching, Schmalz (1998)   (Correct)
This article describes an architecture for the recognition of three-dimensional objects on the basis of viewer centred representations and temporal associations. Considering evidence from psychophysic... / dioeerent kinds of articial neural networks Preprocessing is done by a br by means of an articial neural network is presented by Poggio and

Feedforward Neural Networks for Nonparametric Regression - David Rios Insua, Peter Müller (1998)   (Correct)
Feed forward neural networks (FFNN) with an unconstrained random number of hidden neurons define flexible non-parametric regression models. In M¨uller and Rios Insua (1998) we have argued that variabl... / Opaque this Feedforward neural networks for nonparametric regression br Abstract Feed forward neural networks FFNN with an unconstrained

Bayesian neural networks for classification: how useful is the.. - Penny, Roberts (1998)   (Correct)
This paper presents an empirical assessment of the Bayesian evidence framework for neural networks using four synthetic and four real-world classification problems. We focus on three issues; model sel... / Bayesian neural networks for classification how br evidence framework for neural networks using four synthetic and four

Feature Selection vs Theory Reformulation: a Study of Genetic.. - Burns, Danyluk (1998)   (Correct)
Expert classification systems have proven themselves effective decision makers for many types of problems. However, the accuracy of such systems is often highly dependent upon the accuracy of a human ... / Refinement of Knowledge-based Neural Networks Brendan Davis Burns and br selection knowledgebased neural networks genetic algorithms.

DYNAMO: An Algorithm for Dynamic Acoustic Modeling - Beaufays, Weintraub, Konig (1998)   (Correct)
This paper summarizes part of SRI's effort to improve acoustic modeling in the context of the Large Vocabulary Continuous Speech Recognition (LVCSR) project. It concentrates on two problems that are b... / and by proposing a neural-networkbased architecture to combine br to combine these features. The neural networks NNET are used in

Replicator Equations, Maximal Cliques, and Graph Isomorphism - Pelillo (1998)   (Correct)
We present a new energy-minimization framework for the graph isomorphism problem which is based on an equivalent maximum clique formulation. The approach is centered around a fundamental result proved... / much attention in the neural network community and various br of graphs and images in neural networks J. Phys. A Math. Gen.

A Cascade Neural Network for Blind Signal Extraction without Spurious .. - Thawonmas, CICHOCKI, AMARI (1998)   (Correct)
this paper, we adopt the neural network approach. The main objective of this paper is threefold. 1. To present (in Section 2) a neural network and propose unconstrained extraction and deflation criter... / its Applications A Cascade Neural Network for Blind Signal Extraction br SUMMARY We present a cascade neural network for blind source extraction.

Classification and Pose Estimation of Objects using Nonlinear Features - Talukder, Casasent (1998)   (Correct)
A new nonlinear feature extraction method called the maximum representation and discrimination feature (MRDF) method is presented for extraction of features from input image data. It implements transf... / similar to the Sigma-Pi neural network. However the weights of the br compared to nonlinear neural network implementations. The features

Theory of Neuromata - Síma, Wiedermann (1998)   (Correct)
A finite automaton --- the so-called neuromaton, realized by a finite discrete recurrent neural network, working in parallel computation mode, is considered. Both the size of neuromata (i.e., the numb... / by a finite discrete recurrent neural network working in parallel br Hopfield neuromata i.e.of neural networks with symmetric weights It

Constructive Training of Probabilistic Neural Networks - Berthold, Diamond (1998)   (Correct)
This paper presents an easy to use, constructive training algorithm for Probabilistic Neural Networks, a special type of Radial Basis Function Networks. In contrast to other algorithms, predefinition ... / Training of Probabilistic Neural Networks Michael R. Berthold br algorithm for Probabilistic Neural Networks a special type of Radial

Rotation Invariant Neural Network-Based Face Detection - Rowley, Baluja, Kanade (1998)   (Correct)
In this paper, we present a neural network-based face detection system. Unlike similar systems which are limited to detecting upright, frontal faces, this system detects faces at any degree of rotatio... / Rotation Invariant Neural Network-Based Face Detection br In this paper we present a neural network-based face detection system.

A Visual Interface for Human-Robot Interaction - Heinzmann, Zelinsky (1998)   (Correct)
This paper describes the architecture of a human-robot interaction system and recent work on the vision based human-robot interface. The aim of the project is to develop a robotic system that is safel... / A gaze tracker based on a neural network processing low resolution br gaze tracking using artificial neural networks. Technical Report

Bringing People and Places Together with Dual Augmentation - Mankoff, Somers, Abowd (1998)   (Correct)
This paper describes initial work on the Domisilica project at Georgia Tech. We are exploring the dual augmentation of physical and virtual worlds in Domisilica and applying this novel concept to supp... / of changed behaviors is the Neural Network House Physical spaces br F. C. III and D. Miller. The Neural Network House An overview. Technical

Semantics for using Stochastic Constraint Solvers in Constraint Logic .. - Stuckey, Tam (1998)   (Correct)
F4.729e+05> This paper proposes a number of models for integrating finitedomain stochastic constraint solvers into constraint logic programming systems to solve constraint-satisfaction problems e#ci... / implemented using a modified neural network simulator GENET as a br example simulated annealing neural networks and evolutionary algorithms.

Density Networks with Application to Protein Modelling - Povinelli Churchill (1998)   (Correct)
Density networks are used as probabilistic models of discrete-valued, multi-dimensional data. These models have a neural network architecture, with the outputs corresponding to observed data and the i... / data. These models have a neural network architecture with the outputs br estimation e.g. feed-forward neural networks. These methods work by

Embodied System Life - Pfeifer (1998)   (Correct)
System life is concerned with societies of living beings, i.e. humans, animals, plants, and artifacts, i.e. computers, robots, machines in general, software, and virtual creatures. In this position pa... / test set as for example in neural network applications. There all the br known and it is known how the neural network is embedded in the agent. The

A Common Neural Network Model for Unsupervised Exploratory Data.. - Girolami, Cichocki, Amari (1998)   (Correct)
This paper presents the derivation of an unsupervised learning algorithm, which enables the identification and visualisation of latent structure within ensembles of high dimensional data. This provide... / A Common Neural Network Model for Unsupervised br For I.e.e.e Transactions On Neural Networks Brief Paper Submitted

Dynamic Stochastic Synapses as Computational Units - Maass, Zador (1998)   (Correct)
In most neural network models, synapses are treated as static weights that change only on the slow time scales of learning. It is well known, however, that synapses are highly dynamic, and show usedep... / Abstract In most neural network models synapses are treated br Introduction In most neural network models neurons are viewed as

NP-completeness: A Retrospective - Papadimitriou (1998)   (Correct)
For a quarter of a century now, NP-completeness has been computer science's favorite paradigm, fad, punching bag, buzzword, alibi, and intellectual export. This paper is a fragmentary commentary on ... / database expert neural network and operating system.

Line-Based Face Recognition under Varying Pose - Aeberhard, de Vel (1998)   (Correct)
Most research in recognising human faces consists of full frontal view images and operate under strict imaging conditions such as controlled illumination and limited facial expressions. Face recogniti... / namely templatebased and neural networks. In the template-based br a rotation in depth Neural network-based image techniques use an

Mixture Density Estimation Based on Maximum Likelihood and Sequential .. - Vlassis, Papakonstantinou, Tsanakas (1998)   (Correct)
We address the problem of estimating an unknown probability density function from a sequence of input samples. We approximate the input density with a weighted mixture of a finite number of Gaussian k... / a batch. This is typical for neural network algorithms and may prove br work Recent research in neural networks has proposed several models

Neural Classifier Construction Using Regularization, Pruning and Test .. - Hintz-Madsen, Hansen, Larsen.. (1998)   (Correct)
In this paper we propose a method for construction of feed-forward neural classifiers based on regularization and adaptive architectures. Using a penalized maximum likelihood scheme, we derive a modif... / the Danish Computational Neural Network Center. Jan Larsen thanks the br indeed the objective of most neural network applications. Some of the

Data-Dependent Structural Risk Minimisation for Perceptron Decision.. - Shawe-Taylor, Cristianini (1998)   (Correct)
Perceptron Decision Trees (also known as Linear Machine DTs, etc.) are analysed in order that data-dependent Structural Risk Minimization can be applied. Data-dependent analysis is performed which ind... / approach. Introduction Neural network researchers have traditionally br sigmoid nodes into feedforward neural networks. In this paper we consider a

A Neural Network for the Blind Separation of Non-Gaussian Sources - Freisleben, Hagen, Borschbach (1998)   (Correct)
In this paper, a two--layer neural network is presented that organizes itself to perform blind source separation. The inputs to the network are prewhitened linear mixtures of unknown independent sourc... / A Neural Network for the Blind Separation of br In this paper a two-layer neural network is presented that organizes

Generalized Support Vector Machines - Mangasarian (1998)   (Correct)
By setting apart the two functions of a support vector machine: separation of points by a nonlinear surface in the original space of patterns, and maximizing the distance between separating planes in ... / d ffl Example . Neural Network Kernel AB ab br kernels such as the neural network kernel with a discontinuous

A Bootstrap Evaluation of the Effect of Data Splitting on Financial.. - LeBaron, Weigend (1998)   (Correct)
This article exposes problems of the commonly used technique of splitting the available data into training, validation, and test sets that are held fixed, warns about drawing too strong conclusions fr... / from the data splitting with neural network specific uncertainties br Second on each split the neural network solution with early stopping

Attractor Switching by Neural Control of Chaotic Neurodynamics - Pasemann, Stollenwerk (1998)   (Correct)
Chaotic attractors of discrete-time neural networks include infinitely many unstable periodic orbits, which can be stabilized by small parameter changes in a feedback control. Here we explore the cont... / attractors of discrete-time neural networks include infinitely many br periodic orbits in a chaotic neural network with only two neurons.

Hybrid Interior Point Training of Modular Neural Networks - Szymanski, Lemmon, Bett (1998)   (Correct)
Modular neural networks use a single gating neuron to select the outputs of a collection of agent neurons. Expectation-maximization (EM) algorithms provide one way of training modular neural networks ... / Point Training of Modular Neural Networks Technical Report of the br Point Training of Modular Neural Networks Peter T. Szymanski Michael

User Localisation for Visually-based Human-Machine-Interaction - Boehme, Braumann, Brakensiek.. (1998)   (Correct)
Recently there is an increasing interest in video based interface techniques, allowing more natural interaction between users and systems than common interface devices do. Here, we present a neural ar... / was started to develop a neural network architecture for videobased br Boehme H.J.and Beck C. A Neural Network Hierarchy for Data and

Animat Navigation Using a Cognitive Graph - Trullier, Meyer (1998)   (Correct)
A model of the hippocampus as a "cognitive graph" is proposed. It essentially considers the hippocampus as an heteroassociative network that learns temporal sequences of visited places and stores a to... / who recently proposed a neural network model of short-term memory br of hippocampal function. Neural Networks - .

Rapid Robot Training - Sandra Lee Samelson, Ron Sigal (1998)   (Correct)
this report. In Section 2 we introduce the robot and its graphic simulator, and in Section 3 we give some background on neuro-fuzzy technology. Section 4 presents our training framework, and Section 5... / including artificial neural networks G fuzzy inference systems br we have turned to artificial neural networks where we enjoy the benefits

Neural Network Training and Simulation Using a Multidimensional.. - Likas, Karras, Lagaris (1998)   (Correct)
A new approach is presented to neural network simulation and training that is based on the use of general purpose optimization software. This approach requires that the training problem should be form... / Neural Network Training and Simulation Using br A new approach is presented to neural network simulation and training that

Attribute Grammars for Genetic Representations of Neural Networks and .. - Talib Hussain (1998)   (Correct)
this paper, we give a broad overview of our research into attribute grammar representations, from the basic and known capabilities, to the current ideas being addressed, to the future directions of ou... / for Genetic Representations of Neural Networks and Syntactic Constraints of br is used to specify classes of neural network structures with explicit

The Fluctuations Of The Overlap In The Hopfield Model With Finitely.. - Gentz, Löwe (1998)   (Correct)
We investigate the limiting fluctuations of the order parameter in the Hopfield model of spin glasses and neural networks with finitely many patterns at the critical temperature 1=fi c = 1. At the c... / model of spin glasses and neural networks with finitely many patterns br model of the brain hence as a neural network. This point of view has been

Achieving Robust Behavior by Using Proprioceptive Activity Patterns - Salomon (1998)   (Correct)
This paper proposes a new self-organizing, biologically-inspired control architecture for mobile robots consisting of a controller and a value system. The controller uses activity patterns of visual s... / Mobile Robot Control Neural Networks Introduction One goal br structure is implemented by neural networks. For several reasons

Dynamic on-line clustering and state extraction: An approach to.. - Sreerupa Das (1998)   (Correct)
Introduction Researchers often try to understand the representations that develop in the hidden layers of a neural network during training. Interpretation is difficult because the representations are ... / in the hidden layers of a neural network during training. br In such domains standard neural network learning procedures which

Advanced Methods for Evolutionary Optimisation - Adamidis, Kazarlis, Petridis (1998)   (Correct)
In this paper we present two advanced methods for evolutionary optimisation. One method is based on Parallel Genetic Algorithms. It is called Co-operating Populations with Different Evolution Behaviou... / a Recurrent Artificial Neural Network and the TSP problem and br connected Recurrent Artificial Neural Network RANN to generate two limit

Learning Continuous Attractors in Recurrent Networks - Seung (1998)   (Correct)
One approach to invariant object recognition employs a recurrent neural network as an associative memory. In the standard depiction of the network's state space, memories of objects are stored as attr... / recognition employs a recurrent neural network as an associative memory. In br is to use a recurrent neural network as an associative memory

A Neural Network Model for Prognostic Prediction - Street (1998)   (Correct)
An important and difficult prediction task in many domains, particularly medical decision making, is that of prognosis. Prognosis presents a unique set of problems to a learning system when some of th... / A Neural Network Model for Prognostic br This paper applies artificial neural network classification to the

Combining Time-Delayed Decorrelation And Ica: Towards Solving The.. - Lee, Ziehe, Orglmeister, Sejnowski (1998)   (Correct)
We present methods to separate blindly mixed signals recorded in a room. The learning algorithm is based on the information maximization in a single layer neural network. We focus on the implementatio... / maximization in a single layer neural network. We focus on the br approach in a feedforward neural network implemented in the frequency

Estimating Dependency Structure as a Hidden Variable - Meila, Jordan, Morris (1998)   (Correct)
This paper introduces a probability model, the mixture of trees that can account for sparse, dynamically changing dependence relationships. We present a family of efficient algorithms based on the EM ... / used in the statistics and neural network literature. Of relevance to br and who used multilayer neural networks and knowledge-based neural

Blackjack as a Test Bed for Learning Strategies in Neural Networks - Pérez-Uribe, Sanchez (1998)   (Correct)
Blackjack or twenty-one is a card game where the player attempts to beat the dealer, by obtaining a sum of card values that is equal to or less than 21 so that his total is higher than the dealer's. T... / Joint Conference on Neural Networks IJCNN' Anchorage May br Bed for Learning Strategies in Neural Networks Andr'es P'erez-Uribe and

Hierarchical Evolution of Neural Networks - Moriarty, Miikkulainen (1998)   (Correct)
In most applications of neuro-evolution, each individual in the population represents a complete neural network. Recent work on the SANE system, however, has demonstrated that evolving individual neur... / Hierarchical Evolution of Neural Networks David E. Moriarty y br represents a complete neural network. Recent work on the SANE

Focal-Plane Optical Flow Computation by Foveated CNNs - Marco Balsi (1998)   (Correct)
Optical flow computation is instrumental in robot guidance. Optoelectronic smart-pixel sensors for such computation may be realized on a single chip, by making use of a suitable Cellular Neural Netw... / Fifth Int. Workshop on Cellular Neural Networks and their Applications br use of a suitable Cellular Neural Network architecture defined on a

Neural Optimization - Peterson, Söderberg (1998)   (Correct)
Introduction Many combinatorial optimization problems require a more or less exhaustive search to achieve exact solutions, with the computational effort growing exponentially or worse with system siz... / Handbook of Brain Research and Neural Networks nd edition M.A. Arbib br good solutions. The artificial neural network ANN approach falls within

Including Control Architecture in Attribute Grammar Specifications of .. - Hussain, Browse (1998)   (Correct)
An important problem in evolutionary computing is the design of genetic representations of neural networks that permit optimization of topology and learning characteristics. One promising approach for... / specifications of feedforward neural networks Talib S. Hussain br of genetic representations of neural networks that permit optimization of

A framework for using multiple classifiers in a multiple-agent.. - Vuurpijl, Schomaker (1998)   (Correct)
This paper describes a new framework using intelligent agents for pattern recognition. A justification for using alternatives to current classifier systems is given. The use of the framework, called i... / If trainable statistical or neural-network metaclassifiers are used

A Gradient Descent Method for a Neural Fractal Memory - Melnik, Pollack (1998)   (Correct)
It has been demonstrated that higher order recurrent neural networks exhibit an underlying fractal attractor as an artifact of their dynamics. These fractal attractors offer a very efficent mechanism ... / that higher order recurrent neural networks exhibit an underlying fractal br Keywords-Recurrent Neural Networks Dynamical Systems

Global and Local Neural Network Ensembles. - Sierra, Cruz (1998)   (Correct)
this article a fast Karhunen Lo`eve expansion (Oja (1983)) in a 40 dimensional feature space is used. Following Rumelhart, Hinton and Williams (1986) a multi-layer perceptron with sigmoid maps can be ... / Global and Local Neural Network Ensembles. A. Sierra C. br performance. Key words Neural Network Ensemble Global Neural

Online Learning about Other Agents in a Dynamic Multiagent System - Hu, Wellman (1998)   (Correct)
We analyze the problem of learning about other agents in a class of dynamic multiagent systems, where performance of the primary agent depends on behavior of the others. We consider an online version ... / for s Gammai in his neural network model of the relationship

The role of consciousness and intentionality in perception.. - Bartsch (1998)   (Correct)
This paper discusses the role of consciousness in the distinctions between reception and perception, between a purely causal and a referential or denotational semantics, and between linguistic ability... / connected to the world via a neural network background established in br semantics we also find in neural networks trained towards perception

Using An MDL-Based Cost Function With Neural Networks - Harri Lappalainen (1998)   (Correct)
The minimum description length (MDL) principle is an information theoretically based method to learn models from data. This paper presents how to eOEciently use an MDL-based cost function with neural ... / An Mdl-Based Cost Function With Neural Networks Harri Lappalainen br University of Technology Neural Networks Research Centre P.O.Box

A Hybrid Connectionist and BDI Architecture for Modeling Embedded.. - Kumar (1998)   (Correct)
In this paper, our ongoing work on a hybrid, connectionist and belief-desire-intention (BDI) based, rational agent architecture is described. The architecture makes specific commitments in order to ac... / is a recurrent artificial neural network developed through br to developing intelligent neural network controllers for robots. IEEE

The Evolution of Complexity and the Value of Variability - Anil Seth (1998)   (Correct)
The hypothesis that environmental variability promotes the evolution of organism complexity is explored and illustrated, in two contexts. A coevolutionary `Iterated Prisoner's Dilemma' (IPD) ecology, ... / evolution in simulation of neural network controllers for mobile br and thresholds for a simple neural network to control a Khepera K-Team

Removing Electroencephalographic Artifacts: Comparison between ICA.. - Jung, Humphries, Lee, Makeig.. (1998)   (Correct)
Pervasive electroencephalographic (EEG) artifacts associated with blinks, eye-movements, muscle noise, cardiac signals, and line noise poses a major challenge for EEG interpretation and analysis. Here... / have proposed a simple neural network algorithm that blindly br Proc. IEEE Int'l Conf. on Neural Networks - . Dale

On the use of model order for detecting potential target locations in .. - Kothari, Ensley (1998)   (Correct)
The Region of Interest (ROI) detection stage of an Automatic Target Recognition (ATR) System serves the crucial role of identifying candidate regions which may have potential targets. The large variab... / sigmoidal multi-layered neural networks with lateral connections and br and radial basis function neural networks with a model selection

Automatic Registration of Complex Images Using a Self Organizing.. - Sabisch, Ferguson, Bolouri (1998)   (Correct)
We present a system for automatic mapping of complex gray-scale images onto each other. The system includes a Neocognitron-like structure for hierarchical feature extraction, a 3D Self Organising Map ... / extraction self organising neural networks. I. Introduction A br a combination of artificial neural networks and algorithmic techniques

Network Generating Attribute Grammar Encoding - By Talib (1998)   (Correct)
this document shall make several contributions to the field of neural networks. The primary contributions shall be a new formal framework for specifying neural network architectures, a set of analysis... / Problem The field of neural networks addresses the development br the principles of biological neural networks. Since the resurgence of the

Running Across the Reality Gap: Octopod Locomotion Evolved in a.. - Nick Jakobi (1998)   (Correct)
This paper describes experiments in which neural network control architectures were evolved in minimal simulation for an octopod robot. The robot is around 30cm long and has 4 infra red sensors that... / describes experiments in which neural network control architectures were br the experiments was to evolve neural network control architectures that

Consistency of Posterior Distributions for Neural Networks - Lee (1998)   (Correct)
In this paper we show that the posterior distribution for feedforward neural networks is asymptotically consistent. This paper extends earlier results on universal approximation properties of neural n... / of Posterior Distributions for Neural Networks Herbert Lee May br distribution for feedforward neural networks is asymptotically

Using Independent Component Analysis for Feature Extraction and.. - Andreas Weingessel, Martin Natter.. (1998)   (Correct)
Deriving low-dimensional perceptual spaces from data consisting of many variables is of crucial interest in strategic market planning. A frequently used method in this context is Principal Components ... / computation. Recently neural networks algorithm for extracting br in Section . . . Neural Network Algorithms for PCA Before

Automated Learning and Discovery: State-Of-The-Art and Research.. - Thrun, Faloutsos, Mitchell, Wasserman (1998)   (Correct)
This report summarizes the CONALD meeting, which took place June 11-13, 1998, at Carnegie Mellon University. CONALD brought together an interdisciplinary group of scientists, concerned with decision m... / progression straight neural network discrimination pattern br Trees Boosting Artificial Neural Networks are some of the many

CiteSeer - citeseer.org - Terms of Service - Privacy Policy - Copyright © 1997-2002 NEC Research Institute