A Revisit to Support Vector Data Description (SVDD)
نویسندگان
چکیده
Support vector data description (SVDD), proposed by [1], is a useful method for outlier detection. Its model is obtained by solving the dual optimization problem. In this paper, we point out some issues in their derivations. For example, they formulate SVDD as a non-convex problem and derive the dual problem only under some parameter settings. Given the wide use of SVDD, it is important to address these issues. In this paper, we consider a convex equivalent of SVDD, rigorously derive the dual problem, discuss additional properties, and investigate some novel extensions of SVDD.
منابع مشابه
A Revisit to Support Vector Data Description
Support vector data description (SVDD) is a useful method for outlier detection and has been applied to a variety of applications. However, in the existing optimization procedure of SVDD, there are some issues which may lead to improper usage of SVDD. Some of the issues might already be known in practice, but the theoretical discussion, justification and correction are still lacking. Given the ...
متن کاملIncremental Learning Algorithm for Support Vector Data Description
Support vector data description (SVDD) has become a very attractive kernel method due to its good results in many novelty detection problems.Training SVDD involves solving a constrained convex quadratic programming,which requires large memory and enormous amounts of training time for large-scale data set.In this paper,we analyze the possible changes of support vector set after new samples are a...
متن کاملSampling Method for Fast Training of Support Vector Data Description
Support Vector Data Description (SVDD) is a machine learning technique used for single class classification and outlier detection. The SVDD model for normal data description builds a minimum radius hypersphere around the training data. A flexible description can be obtained by use of Kernel functions. The data description is defined by the support vectors obtained by solving quadratic optimizat...
متن کاملEllipse Support Vector Data Description
This paper presents a novel Boundary-based approach in one-class classification that is inspired by support vector data description (SVDD). The SVDD is a popular kernel method which tries to fit a hypersphere around the target objects and of course more precise boundary is relied on selecting proper parameters for the kernel functions. Even with a flexible Gaussian kernel function, the SVDD cou...
متن کاملData domain description using support vectors
This paper introduces a new method for data domain description , inspired by the Support Vector Machine by V.Vapnik, called the Support Vector Domain Description SVDD. This method computes a sphere shaped decision boundary with minimal volume around a set of objects. This data description can be used for novelty or outlier detection. It contains support vectors describing the sphere boundary an...
متن کامل