Y
Yoshiki Suzuki
Researcher at Nagoya Institute of Technology
Publications - 8
Citations - 159
Yoshiki Suzuki is an academic researcher from Nagoya Institute of Technology. The author has contributed to research in topics: Regularization (mathematics) & Support vector machine. The author has an hindex of 5, co-authored 8 publications receiving 146 citations.
Papers
More filters
Proceedings Article
Safe Screening of Non-Support Vectors in Pathwise SVM Computation
TL;DR: It is claimed that some of the nonsupport vectors (non-SVs) that have no influence on the SVM classifier can be screened out prior to the training phase in pathwise SVM computation scenario, in which one is asked to train a sequence of S VM classifiers for different regularization parameters.
Posted Content
Quick sensitivity analysis for incremental data modification and its application to leave-one-out CV in linear classification problems
TL;DR: In this paper, the authors propose a framework for large scale classification problems that can be used when a small number of instances are incrementally added or removed, without actually re-optimizing the original training set.
Proceedings ArticleDOI
Quick Sensitivity Analysis for Incremental Data Modification and Its Application to Leave-one-out CV in Linear Classification Problems
TL;DR: This paper introduces a novel sensitivity analysis framework that can quickly provide a lower and an upper bounds of a quantity on the unknown updated classifier and demonstrates that the bounds provided by the framework are often sufficiently tight for making desired inferences.
Posted Content
Safe Sample Screening for Support Vector Machines
TL;DR: This paper introduces a new approach called safe sample screening that enables us to identify a subset of the non-SVs and screen them out prior to the training phase, and proves that it can substantially decrease the computational cost of the state-of-the-art SVM solvers such as LIBSVM.
Proceedings Article
Regularization path of cross-validation error lower bounds
TL;DR: In this paper, the authors propose a framework for computing a lower bound of the CV errors as a function of the regularization parameter, which they call regularization path of CV error lower bounds.