Weighted Classification Cascades for Optimizing Discovery Significance in the HiggsML Challenge

Journal Article

We introduce a minorization-maximization approach to optimizing common measures of discovery significance in high energy physics. The approach alternates between solving a weighted binary classification problem and updating class weights in a simple, closed-form manner. Moreover, an argument based on convex duality shows that an improvement in weighted classification error on any round yields a commensurate improvement in discovery significance. We complement our derivation with experimental results from the 2014 Higgs boson machine learning challenge.

Full Text

Duke Authors

Cited Authors

  • Mackey, L; Bryan, J; Mo, MY

Cited Editors

  • Cowan, G; Germain, C; Guyon, I; Kégl, B; Rousseau, D

Published Date

  • December 13, 2015

Published In

  • Proceedings of the Nips 2014 Workshop on High Energy Physics and Machine Learning

Volume / Issue

  • 42 /

Start / End Page

  • 129 - 134

Published By