Working Set Selection Using Second Order Information for Training Svm, " Complexity-reduced Scheme for Feature Extraction with Linear Discriminant Analysis

1 2 (w · w) + γ 2 (w * · w *) + C n i=1 ξ * i s.t. y i ((w · z i) + b) ≥ 1 − ξ The dual optimization problem of (29) is minimize α − n i=1 α i + 1 2 n i, j =1 α i α j y i y j (z i · z j) + 1 2γ n i, j =1 (α i + β i − C)(α j + β j − C)(z * i · z * j) s.t. ACKNOWLEDGMENT The authors would like to thank the anonymous reviewers for their helpful technical feedback. REFERENCES [1] B. Bakker and T. Heskes, " Task clustering and gating for Bayesian multitask learning, " A framework for learning predictive structures from multiple tasks and unlabeled data, " [20] J. Platt, " Using sparseness and analytic QP to speed training of support vector machines, " in Multi-task learning for classification with Dirichlet process priors, " Abstract— Owing to the singularity of the within-class scatter, linear discriminant analysis (LDA) becomes ill-posed for small sample size (SSS) problems. Null-space-based LDA (NLDA), which is an extension of LDA, provides good discriminant performances for SSS problems. Yet, as the original scheme for the feature extractor (FE) of NLDA suffers from a complexity burden, a few modified schemes have since been proposed for complexity reduction. In this brief, by transforming the problem of finding the FE of NLDA into a linear equation problem, a novel scheme is derived, offering a further reduction of the complexity.