Performance and optimization of support vector machines in high-energy physics classification problems

详细信息    查看全文

• 作者：M.Ö ; . Sahin ; ^{ozgur.sahin@desy.de" class="auth_mail" title="E-mail the corresponding author} ; D. Krü ; cker ^{dirk.kruecker@desy.de" class="auth_mail" title="E-mail the corresponding author} ; I.-A. Melzer-Pellmann ^{isabell.melzer@desy.de" class="auth_mail" title="E-mail the corresponding author}
• 关键词：Support vector machine ; Multivariate analysis ; Beyond the standard model search ; High energy physics ; Supersymmetry
• 刊名：Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
• 出版年：2016
• 出版时间：1 December 2016
• 年：2016
• 卷：838
• 期：Complete
• 页码：137-146
• 全文大小：1523 K

文摘

In this paper we promote the use of Support Vector Machines (SVM) as a machine learning tool for searches in high-energy physics. As an example for a new-physics search we discuss the popular case of Supersymmetry at the Large Hadron Collider. We demonstrate that the SVM is a valuable tool and show that an automated discovery-significance based optimization of the SVM hyper-parameters is a highly efficient way to prepare an SVM for such applications.