• No results found

robust neural network classifiers

Robust Neural Network Classifier

Robust Neural Network Classifier

... classes. Neural Networks (NN) are an effective tool in the field of pattern ...training neural networks. However (MSE) based learning algorithm is not robust in presence of outliers that may pollute ...

6

Sparse Robust Regression for Explaining Classifiers

Sparse Robust Regression for Explaining Classifiers

... Explaining internal layers Since slise only requires the data to be in vector- form it is easy to combine slise with other methods. Of the categories from Section 2.2.2 the model inspection methods are of particular ...

68

Lower bounds on the robustness to adversarial perturbations

Lower bounds on the robustness to adversarial perturbations

... state-of-the-art neural networks are sig- nificantly ...a neural network used for image recognition to misclassify its input by applying very specific, hardly perceptible perturbations to the input, ...

10

Neural Network Classifiers for Human Tissue Classification in NIR Biomedical Multispectral Imaging

Neural Network Classifiers for Human Tissue Classification in NIR Biomedical Multispectral Imaging

... Mendenhall et al. approach skin classification by using both the visible and NIR spec- trum, as many approaches that solely rely on the colour spectrum can have relatively high false-alarm rates (8% to 15%) [17]. The ...

73

Comparative Study of Classification Techniques on Breast Cancer FNA Biopsy Data

Comparative Study of Classification Techniques on Breast Cancer FNA Biopsy Data

... Bayesian network classifier uses the unsupervised learning algorithm, where the class target is unknown though we have the inputs (attributes) [14] and the classifier learning algorithm can be structured into two ...

8

Comparison of Adaboost and Bagging Ensemble Method for Prediction of Heart Disease

Comparison of Adaboost and Bagging Ensemble Method for Prediction of Heart Disease

... Data mining is becoming one of the most indispensable and inspiring area of research because of its ability to discover significant information from a pool of data ([SS16]). The extracted information can be usefully ...

12

Automatic classification of field-collected dinoflagellates by artificial neural network

Automatic classification of field-collected dinoflagellates by artificial neural network

... The performance by human 'expert' ecologists/taxonon~ists in identifying these species was compared to that achieved by 2 art~fi- cial neural network classifiers (multilay[r] ...

7

GLOBAL JOURNAL OF ADVANCED ENGINEERING TECHNOLOGIES AND SCIENCES RECOGNITION OF PERSIAN HANDWRITTEN NUMBERS BASED ON ASSEMBLY OF REINFORCED CLASSIFIERS Hamid Parvin*, Seyed Ahad Zolfagharifar, Faramarz Karamizadeh

GLOBAL JOURNAL OF ADVANCED ENGINEERING TECHNOLOGIES AND SCIENCES RECOGNITION OF PERSIAN HANDWRITTEN NUMBERS BASED ON ASSEMBLY OF REINFORCED CLASSIFIERS Hamid Parvin*, Seyed Ahad Zolfagharifar, Faramarz Karamizadeh

... binary classifiers is used. If we consider the numbers of the classifiers as c, this method needs to use (c×(c-1))/2 classifiers ...binary classifiers have a significantly ...

11

A Convolutional Neural Network Model Robust To Distorted Fingerprints

A Convolutional Neural Network Model Robust To Distorted Fingerprints

... Abstract: The greatest challenge in fingerprint recognition is verifying distorted fingerprints. Distortion in fingerprints may arise from errors introduced while acquiring fingerprints, the nature of the fingerprints or ...

6

Deep residual neural network for EMI event classification using bispectrum representations

Deep residual neural network for EMI event classification using bispectrum representations

... residual network differs from a standard Convolutional Neural Network (CNN) through the implementation of identity skip connec- tions or ...residual network utilised in this work is based on ...

5

Multiple Fault Detection in a Four Stroke Engine Using Single Sensor System

Multiple Fault Detection in a Four Stroke Engine Using Single Sensor System

... a neural network increases, the complexity of computation is also seen to ...the network was designed by keeping a number of Hidden layer #1 (L1) PEs fixed to 5 and by varying Hidden layer #2 (L2) ...

11

Class Disjointness Constraints as Specific Objective Functions in Neural Network Classifiers

Class Disjointness Constraints as Specific Objective Functions in Neural Network Classifiers

... Increasing performance of deep learning techniques on computer vision tasks like object detec- tion has led to systems able to detect a large number of classes of objects. Most deep learning models use simple ...

8

Knowledge Extraction Using Probabilistic Reasoning: An Artificial Neural Network Approach

Knowledge Extraction Using Probabilistic Reasoning: An Artificial Neural Network Approach

... artificial neural network algorithms, a successfully developed prototype has been developed that demonstrates the applicability of the ...random neural network classifier (RNNC) performed ...

8

Robust Face Recognition Based on Convolutional Neural Network

Robust Face Recognition Based on Convolutional Neural Network

... Convolutional Neural Network model, and the center loss layer jointly to enhance the discriminative of the designed network ...The network is trained on the self-expanding CASIA-WebFace ...

6

A Constrained Multi-Objective Learning Algorithm for Feed-Forward Neural Network Classifiers

A Constrained Multi-Objective Learning Algorithm for Feed-Forward Neural Network Classifiers

... and neural networks [2, ...Feed-forward Neural Network (FNN) based classifiers for solving two-class classification tasks, that is, classifiers that assign patterns to one of two ...

9

Integrating Predictions from Neural Network Relation Classifiers into Coreference and Bridging Resolution

Integrating Predictions from Neural Network Relation Classifiers into Coreference and Bridging Resolution

... state-of-the-art neural-network classifiers to predict semantic relations between noun pairs, and integrate the relation predictions into existing systems for coreference and bridging ...

6

Detection of sodium oxalate needles in optical images using neural network classifiers

Detection of sodium oxalate needles in optical images using neural network classifiers

... If a General Regression Neural Network (GRNN) [4,5] is used as a classifier the number of training data points is taken to be in proportion with the a priori probability of occurre[r] ...

5

A Study of Genetic Neural Network as Classifiers and its Application in Breast Cancer Diagnosis

A Study of Genetic Neural Network as Classifiers and its Application in Breast Cancer Diagnosis

... Artificial neural networks, when equipped with the ability of self-organizing and self-learning with high stability can process the nonlinear problem in the identification of the benign and malignant cells ...

6

Online isolated handwriting and text recognition based on annotated image features

Online isolated handwriting and text recognition based on annotated image features

... The representation schemes of input pattern and model database are of particular importance since a classification method depends largely on them (Liu et al., 2004). Selecting the data representation is one of the most ...

36

Sparse Robust Regression for Explaining Classifiers

Sparse Robust Regression for Explaining Classifiers

... are robust to outliers. In this paper we develop a robust regression method for finding the largest subset in the data that can be approximated using a sparse linear model to a given ...state-of-the-art ...

16

Show all 10000 documents...

Related subjects