Weighted Classification Cascades for Optimizing Discovery Significance in the HiggsML Challenge
This work addresses the specific problem of enhancing discovery significance in high-energy physics experiments, representing an incremental improvement over existing methods.
The paper tackled the problem of optimizing discovery significance in high energy physics by introducing a minorization-maximization approach that alternates between solving weighted binary classification and updating weights, resulting in improved performance as demonstrated in the 2014 Higgs boson machine learning challenge.
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.