Global Plus Local Jointly Regularized Support Vector Data Description for Novelty Detection

Hu, Wenjun; Hu, Tianjie; Wei, Yuzhen; Lou, Jungang; Wang, Shitong

doi:10.1109/tnnls.2021.3129321

Cited by 14 publications

(5 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By its part, the domain-based techniques include the Support Vector Data Description [154] and the One Class Support Vector Machine [155]. These methods describe a domain that have normal data, also define the limits that round the normal class and that follows the distribution of the data, but they do not provide an explicit distribution of the regions with high density.…”

Section: Techniques That Could Be Possible Potential Solutions To The...mentioning

confidence: 99%

Advances in Power Quality Analysis Techniques for Electrical Machines and Drives: A Review

et al. 2022

View full text Add to dashboard Cite

The electric machines are the elements most used at an industry level, and they represent the major power consumption of the productive processes. Particularly speaking, among all electric machines, the motors and their drives play a key role since they literally allow the motion interchange in the industrial processes; it could be said that they are the medullar column for moving the rest of the mechanical parts. Hence, their proper operation must be guaranteed in order to raise, as much as possible, their efficiency, and, as consequence, bring out the economic benefits. This review presents a general overview of the reported works that address the efficiency topic in motors and drives and in the power quality of the electric grid. This study speaks about the relationship existing between the motors and drives that induces electric disturbances into the grid, affecting its power quality, and also how these power disturbances present in the electrical network adversely affect, in turn, the motors and drives. In addition, the reported techniques that tackle the detection, classification, and mitigations of power quality disturbances are discussed. Additionally, several works are reviewed in order to present the panorama that show the evolution and advances in the techniques and tendencies in both senses: motors and drives affecting the power source quality and the power quality disturbances affecting the efficiency of motors and drives. A discussion of trends in techniques and future work about power quality analysis from the motors and drives efficiency viewpoint is provided. Finally, some prompts are made about alternative methods that could help in overcome the gaps until now detected in the reported approaches referring to the detection, classification and mitigation of power disturbances with views toward the improvement of the efficiency of motors and drives.

show abstract

Section: Techniques That Could Be Possible Potential Solutions To The...mentioning

confidence: 99%

Advances in Power Quality Analysis Techniques for Electrical Machines and Drives: A Review

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Due to its success in data description and its intuitive geometrical interpretation and the ability to benefit from a kernel-based representation, the SVDD approach serves as a widely used technique in the OCC literature, motivating many subsequent research. As an instance, in [35], based on the observation that the SVDD centre and the volume are sensitive to the parameter controlling the trade-off between the errors (slacks) and the volume, a method called GL-SVDD is proposed where local and global probability densities are used to derive sample-adaptive errors via associating weights to the slacks corresponding to different objects. In [36], a different sample-specific weighting approach (P-SVDD) based on the position of the feature space image is proposed to adaptively regularise the complexity of the SVDD sphere.…”

Section: Prior Workmentioning

confidence: 99%

“…The proposed approach is denoted as " p -SVDD" in the corresponding tables. We also provide a comparison of the proposed p -SVDD method to some linear re-weighting variants of the SVDD approach including P-SVDD [36], DW-SVDD [37], and GL-SVDD [35] as well as state-of-the-art OCC techniques. In particular, we have included kernel-based one-class classifiers which are applicable to moderately-sized datasets.…”

Section: Implementation Detailsmentioning

confidence: 99%

“…Furthermore, although the 2 -SVDD method provides some improvement with respect to the original 1 -SVDD approach, its performance is still inferior compared to the proposed method. The second best performing method (in terms of average ranking) corresponds to a sample reweighting SVDD approach presented in [35] which uses global and local statistics to linearly weight slacks in the objective function.…”

Section: Pure One-class Settingmentioning

confidence: 99%

See 1 more Smart Citation

$\ell_p$ Slack Norm Support Vector Data Description

Arashloo¹

2022

Preprint

View full text Add to dashboard Cite

The support vector data description (SVDD) approach serves as a de facto standard for one-class classification where the learning task entails inferring the smallest hyper-sphere to enclose target objects while linearly penalising any errors/slacks via an 1 -norm penalty term. In this study, we generalise this modelling formalism to a general p -norm (p ≥ 1) slack penalty function. By virtue of an p slack norm, the proposed approach enables formulating a nonlinear cost function with respect to slacks. From a dual problem perspective, the proposed method introduces a sparsity-inducing dual norm into the objective function, and thus, possesses a higher capacity to tune into the inherent sparsity of the problem for enhanced descriptive capability.A theoretical analysis based on Rademacher complexities characterises the generalisation performance of the proposed approach in terms of parameter p while the experimental results on several datasets confirm the merits of the proposed method compared to other alternatives.

show abstract

“…Inspired by support vector classifier, support vector data description (SVDD) (Tax & Duin, 1999) characterizes a data set by obtaining the spherically shaped boundary. Through a model built to describe the target data set, it benefits a wide range of applications, such as image description (Aslani & Seipel, 2021), novelty discovery (Hu et al, 2023), adversarial training (C. , and machine unlearning (M. . However, in collecting support vectors (SVs) for data description, the conventional solution conducts model training through solving a quadratic programming optimization problem.…”

Section: Introductionmentioning

confidence: 99%

Selecting Indispensable Edge Patterns With Adaptive Sampling and Double Local Analysis for Data Description

Li,

Ping

2024

Journal of Cases on Information Technology

View full text Add to dashboard Cite

Support vector data description (SVDD) inspires us in data analysis, adversarial training, and machine unlearning. However, collecting support vectors requires pricey computation, while the alternative boundary selection with O(N2) is still a challenge. The authors propose an indispensable edge pattern selection method (IEPS) for data description with direct SVDD model building. IEPS suggests a double local analysis to select the global edge patterns. Edge patterns belong to a subset of the target problem of SVDD and its variants, and neighbor analysis becomes pivotal. While an excessive number of participating data result in redundant computations, an insufficient number may impede data separability or compromise the model's quality. Consequently, a data-adaptive sampling strategy has been devised to ascertain an optimal ratio of retained data for edge pattern selection. Extensive experiments indicate that IEPS keeps indispensable edge patterns for data description while reducing the interference in the norm vector generation to guarantee the effectiveness for clustering analysis.

show abstract

Global Plus Local Jointly Regularized Support Vector Data Description for Novelty Detection

Cited by 14 publications

References 45 publications

Advances in Power Quality Analysis Techniques for Electrical Machines and Drives: A Review

Advances in Power Quality Analysis Techniques for Electrical Machines and Drives: A Review

$\ell_p$ Slack Norm Support Vector Data Description

Selecting Indispensable Edge Patterns With Adaptive Sampling and Double Local Analysis for Data Description

Contact Info

Product

Resources

About