Automatic pattern classifiers that allow for on-line incremental learning can adapt internal class models efficiently in response to new information without retraining from the start using all training data and withou...
详细信息
ISBN:
(纸本)9783540699385
Automatic pattern classifiers that allow for on-line incremental learning can adapt internal class models efficiently in response to new information without retraining from the start using all training data and without being subject to catastrophic forgeting. In this paper, the performance of the fuzzy ARTMAP neural network for supervised incremental learning is compared to that of supervised batch learning. An experimental protocole is presented to assess this network's potential for incremental learning of new blocks of training data, in terms of generalization error and resource requirements, using several synthetic patternrecognition problems. The advantages and drawbacks of training fuzzy ARTMAP incrementally are assessed for different data block sizes and data set structures. Overall results indicate that error rate of fuzzy ARTMAP is significantly higher when it is trained through incremental learning than through batch learning. As the size of training blocs decreases, the error rate acheived through incremental learning grows, but provides a more compact network using fewer training epochs. In the cases where the class distributions overlap, incremental learning shows signs of over-training. With a growing numbers of training patterns, the error rate grows while the compression reaches a plateau.
In this paper, we propose an iris recognition method based on genetic algorithms (GA) to select the optimal features subset. The iris data usually contains huge number of textural features and a comparatively small nu...
详细信息
ISBN:
(纸本)9783540699385
In this paper, we propose an iris recognition method based on genetic algorithms (GA) to select the optimal features subset. The iris data usually contains huge number of textural features and a comparatively small number of samples per subject, which make the accurate iris patterns classification challenging. Feature selection scheme is used to identify the most important and irrelevant features from extracted features set of relatively high dimension based on some selection criterions. The traditional feature selection schemes require sufficient number of samples per subject to select the most representative features sequence;however, it is not always practical to accumulate a large number of samples due to some security issues. In this paper, we propose GA to improve the feature subset selection by combining valuable outcomes from multiple feature selection methods. The main objective of GA is to achieve a balance among the recognition rate, the false accept rate, the false reject rate and the selected features subset size. This paper also motivates and introduces the use of Gaussian Mixture Model for iris pattern classification. The proposed technique is computationally effective with the recognition rates of 97.81% and 96.23% on the ICE (Iris Challenge Evaluation) and the WVU (West Virginia University) iris datasets respectively.
In this paper, a perception-based algorithm for fusion of multiple fingerprint matchers is presented. The person to be identified submits to the personal authentication system her/his fingerprint and claimed identity....
详细信息
In this paper, a perception-based algorithm for fusion of multiple fingerprint matchers is presented. The person to be identified submits to the personal authentication system her/his fingerprint and claimed identity. Multiple fingerprint matchers provide a set of verification scores, that are then fused by a single-layer perceptron with class-separation loss function. Weights of such perceptron are explicitly optimised to increase the separation between genuine users and impostors (i.e., unauthorized users). Reported experiments show that such modified perceptron allows improving the performances and the robustness of the best individual fingerprint matcher, and outperforming some commonly used fusion rules. (c) 2005 Elsevier B.V. All rights reserved.
neural network models for unsupervised patternrecognition learning are challenged when the difference between the patterns of the training set is small. The standardneural network architecture for pattern recognitio...
详细信息
ISBN:
(纸本)3540379517
neural network models for unsupervised patternrecognition learning are challenged when the difference between the patterns of the training set is small. The standardneural network architecture for patternrecognition learning consists of adaptive forward connections. and lateral inhibition, which provides competition between output neurons. We propose an additional adaptive inhibitory feedback mechanism, to emphasize the difference between training patterns and improve learning. We present an implementation of adaptive feedback inhibition for spiking neural network models, based on spike timing dependent plasticity (STDP). When the inhibitory feedback connections are adjusted using an anti-Hebbian learning rule, feedback inhibition suppresses the redundant activity of input units which code the overlap between similar stimuli. We show, that learning speed and pattern discriminatability can be increased by adding this mechanism to the standard architecture.
Diagnosis of disease based on the classification of DNA microarray gene expression profiles of clinical samples is a promising novel approach to improve the performance and accuracy of current routine diagnostic proce...
详细信息
ISBN:
(纸本)9783540699385
Diagnosis of disease based on the classification of DNA microarray gene expression profiles of clinical samples is a promising novel approach to improve the performance and accuracy of current routine diagnostic procedures. In many applications ensembles outperform single classifiers. In a clinical setting a combination of simple classification rules, such as single threshold classifiers on individual gene expression values, may provide valuable insights and facilitate the diagnostic process. A boosting algorithm can be used for building such decision rules by utilizing single threshold classifiers as base classifiers. AdaBoost can be seen as the predecessor of many boosting algorithms developed, unfortunately its performance degrades on high-dimensional data. Here we compare extensions of AdaBoost namely MultiBoost, MadaBoost and AdaBoost-VC in cross-validation experiments on noisy high-dimensional artifical and real data sets. The artifical data sets are so constructed, that features, which are relevant for the class distinction, can easily be read out. Our special interest is in the features the ensembles select for classification and how many of them are effectively related to the original class distinction.
Recently, a number of authors have explored the use of re- cursive recursive neural nets (RNN) for the adaptive processing of trees or tree-like structures. One of the most important language-theoretical formalization...
详细信息
Automatic eye tracking is a challenging task, with numerous applications in biometrics, security, intelligent human-computer interfaces;and driver's sleepiness detection systems. Eye localization and extraction is...
详细信息
ISBN:
(纸本)3540379517
Automatic eye tracking is a challenging task, with numerous applications in biometrics, security, intelligent human-computer interfaces;and driver's sleepiness detection systems. Eye localization and extraction is, therefore, the first step to the solution of such problems. In this paper, we present a new method, based on neural autoassociators, to solve the problem of detecting eyes from a facial image. A subset of the AR Database, collecting individuals both with or without glasses and with open or closed eyes, has been used for experiments and benchmarking. Preliminary experimental results are very promising and demonstrate the efficiency of the proposed eye localization system.
We present in this paper a new facial feature localizer. It uses a kind of auto-associative neural network trained to localize specific facial features (like eyes and mouth corners) in orientation-free faces. One poss...
详细信息
ISBN:
(纸本)3540379517
We present in this paper a new facial feature localizer. It uses a kind of auto-associative neural network trained to localize specific facial features (like eyes and mouth corners) in orientation-free faces. One possible extension is presented where several specialized detectors are trained to deal with each face orientation. To select the best localization hypothesis, we combine radiometric and probabilistic information. The method is quite fast and accurate. The mean localization error (estimated on more than 700 test images) is lower than 9%.
In this paper, we present experiments comparing different training algorithms for Radial Basis Functions (RBF) neuralnetworks. In particular we compare the classical training which consist of an unsupervised training...
详细信息
ISBN:
(纸本)3540379517
In this paper, we present experiments comparing different training algorithms for Radial Basis Functions (RBF) neuralnetworks. In particular we compare the classical training which consist of an unsupervised training of centers followed by a supervised training of the weights at the output, with the full supervised training by gradient descent proposed recently in same papers. We conclude that a fully supervised training performs generally better. We also compare Batch training with Online training and we conclude that Online training suppose a reduction in the number of iterations.
Mathematical models are normally used to calculate the component concentrations in biological wastewater treatment. However, this work deals with the wastewater from a coke plant and it implies inhibition effects betw...
详细信息
ISBN:
(纸本)3540379517
Mathematical models are normally used to calculate the component concentrations in biological wastewater treatment. However, this work deals with the wastewater from a coke plant and it implies inhibition effects between components which do not permit the use of said mathematical models. Due to this, feed-forwardneuralnetworks were used to estimate the ammonium concentration in the effluent stream of the biological plant. The architecture of the neural network is based on previous works in this topic. The methodology consists in performing a group of different sizes of the hidden layer and different subsets of input variables.
暂无评论