Weighted Classification Cascades for Optimizing Discovery Significance in the HiggsML Challenge

2014/09/09 by Lester Mackey, Mackey, Lester, Jordan Bryan +3
Computer Science · Mathematics · #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #cs.LG #stat.ML

paper · pdf · doi:10.48550/arxiv.1409.2655

arxiv created 2015/09/10 · arxiv updated 2015/09/11

Abstract

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.

Related

• Extracting more from boosted decision trees: A high energy physics case study
• The Higgs Machine Learning Challenge
• Optimal event selection and categorization in high energy physics, Part 1: Signal discovery
• A simple guide from Machine Learning outputs to statistical criteria
• Evaluating Modifications to Classifiers for Identification of Higgs Bosons