Support Vector Method for Novelty Detection
Bernhard Schölkopf, Robert C. Williamson, Alex J. Smola et al. · 1999
Suppose you are given some dataset drawn from an underlying probability distributionPand you want to estimate a “simple ” subsetSof input space such that the probability that a test point drawn from P lies outside of Sequals some a priori specified between0and1. We propose a m ethod to approach this problem by trying to estimate a function f which is positive on S and negative on the complement. The functional form offis given by a kernel expansion in terms of a potentially small subset of the training data; it is regularized by controlling the length of the weight vector in an associated feature space. We provide a theoretical analysis of the statistical performance of our algorithm. The algorithm is a natural extension of the support vector algorithm to the case of unlabelled data.
Cited by 1,997 works.
Authors
References
- Statistical Learning Theory
- Estimating the Support of a High-Dimensional Distribution
- Extracting Support Data for a Given Task
- Pattern recognition using generalized portrait method
- Kernel principal component analysis
- Regularized Principal Manifolds
- Learning Distributions by Their Density Levels: A Paradigm for Learning without a Teacher
- Single-class Support Vector Machines