A null space based one class kernel Fisher discriminant

Dufrenois, Franck; Noyer, Jean-Charles

doi:10.1109/ijcnn.2016.7727608

Cited by 7 publications

(11 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since novel examples do not exist prior to test time, training is carried out using oneclass learning principles. In the previous works [4], [11], the performance of novelty detection has been assessed based on different classes of the ImageNet and the Caltech 256 datasets. Since all CNNs used in our work have been trained using the ImageNet dataset, we use the Caltech 256 dataset to evaluate the performance of one-class novelty detection.…”

Section: B Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Learning Deep Features for One-Class Classification

Perera

Patel

2019

IEEE Trans. on Image Process.

330

187

View full text Add to dashboard Cite

We present a novel deep-learning based approach for one-class transfer learning in which labeled data from an unrelated task is used for feature learning in one-class classification. The proposed method operates on top of a Convolutional Neural Network (CNN) of choice and produces descriptive features while maintaining a low intra-class variance in the feature space for the given class. For this purpose two loss functions, compactness loss and descriptiveness loss are proposed along with a parallel CNN architecture. A template matching-based framework is introduced to facilitate the testing process. Extensive experiments on publicly available anomaly detection, novelty detection and mobile active authentication datasets show that the proposed Deep One-Class (DOC) classification method achieves significant improvements over the state-of-the-art.

show abstract

Section: B Resultsmentioning

confidence: 99%

“…Dev.) One Class SVM [43] 0.606 (0.003) KNFST [4] 0.575 (0.004) Oc-KNFD [11] 0.619 (0.003) Autoencoder [15] 0.532(0.003) OCNN AlexNet [5] 0.907 (0.029) OCNN VGG16 [5] 0 classes. Therefore, in this particular case, ImageNet samples are a good representative of novel object classes present in Caltech 256.…”

Section: B Resultsmentioning

confidence: 99%

Learning Deep Features for One-Class Classification

Perera

Patel

2019

IEEE Trans. on Image Process.

330

187

View full text Add to dashboard Cite

show abstract

“…Another drawback is that the method is based on the assumption that the target population differs from the outlier population in terms of their respective densities which might not hold for real-world problems in general. A later study [45] tries to address these shortcomings via a null-space variant of the method in [39], [52]. In order to overcome the limitation of the availability of outlier samples, it is proposed to separate the target class from the origin of the kernel feature space, which serves as an artificial outlier sample.…”

Section: Related Workmentioning

confidence: 99%

“…Nevertheless, the high computational cost associated with these methods can be considered as a bottleneck in their usage. For instance, the one-class variants of kernel discriminant analysis [38], [45], [39], [46] often require computationally intensive eigendecompositions of dense matrices.…”

Section: Introductionmentioning

confidence: 99%

Robust One-Class Kernel Spectral Regression

Arashloo

Kittler

2021

IEEE Trans. Neural Netw. Learning Syst.

View full text Add to dashboard Cite

The kernel null-space technique and its regression-based formulation (called one-class kernel spectral regression, a.k.a. OC-KSR) is known to be an effective and computationally attractive one-class classification framework. Despite its outstanding performance, the applicability of kernel null-space method is limited due to its susceptibility to possible training data corruptions and inability to rank training observations according to their conformity with the model. This work addresses these shortcomings by studying the effect of regularising the solution of the null-space kernel Fisher methodology in the context of its regression-based formulation (OC-KSR). In this respect, first, the effect of a Tikhonov regularisation in the Hilbert space is analysed where the one-class learning problem in presence of contaminations in the training set is posed as a sensitivity analysis problem. Next, driven by the success of the sparse representation methodology, the effect of a sparsity regularisation on the solution is studied. For both alternative regularisation schemes, iterative algorithms are proposed which recursively update label confidences and rank training observations based on their fit with the model. Through extensive experiments conducted on different data sets, the proposed methodology is found to enhance robustness against contamination in the training set as compared with the baseline kernel null-space technique as well as other existing approaches in a one-class classification paradigm while providing the functionality to rank training samples effectively.

show abstract

“…from multi-class discriminant analysis approaches dealing with this small sample size problem [11] indicates that null space directions contain high discrimination power [12], [13], [14]. Interestingly, one-class discrimination approaches based on null space analysis have been also recently proposed [15], [16]. However, the latter ones are in fact designed by following a multi-class setting and cannot be directly extended for classspecific discrimination.…”

Section: Introductionmentioning

confidence: 99%

Null Space Analysis for Class-Specific Discriminant Learning

Raitoharju,

Iosifidis

2019

Preprint

View full text Add to dashboard Cite

In this paper, we carry out null space analysis for Class-Specific Discriminant Analysis (CSDA) and formulate a number of solutions based on the analysis. We analyze both theoretically and experimentally the significance of each algorithmic step. The innate subspace dimensionality resulting from the proposed solutions is typically quite high and we discuss how the need for further dimensionality reduction changes the situation. Experimental evaluation of the proposed solutions shows that the straightforward extension of null space analysis approaches to the class-specific setting can outperform the standard CSDA method. Furthermore, by exploiting a recently proposed outof-class scatter definition encoding the multi-modality of the negative class naturally appearing in class-specific problems, null space projections can lead to a performance comparable to or outperforming the most recent CSDA methods.

show abstract

A null space based one class kernel Fisher discriminant

Cited by 7 publications

References 20 publications

Learning Deep Features for One-Class Classification

Learning Deep Features for One-Class Classification

Robust One-Class Kernel Spectral Regression

Null Space Analysis for Class-Specific Discriminant Learning

Contact Info

Product

Resources

About