Punzi-loss: A non-differentiable metric approximation for sensitivity optimisation in the search for new particles

TL;DR

This paper introduces the Punzi-loss, a novel non-differentiable metric approximation and loss-scheduling method for particle search, enabling a single neural network to effectively classify signals across multiple mass hypotheses.

Contribution

The paper presents the Punzi-loss function and Punzi-net, a new approach that outperforms standard methods and generalizes across different mass hypotheses in particle physics searches.

Findings

Punzi-net outperforms standard multivariate analysis techniques.

It generalizes well to unseen mass hypotheses.

The approach is implemented in PyTorch with publicly available code.

Abstract

We present the novel implementation of a non-differentiable metric approximation and a corresponding loss-scheduling aimed at the search for new particles of unknown mass in high energy physics experiments. We call the loss-scheduling, based on the minimisation of a figure-of-merit related function typical of particle physics, a Punzi-loss function, and the neural network that utilises this loss function a Punzi-net. We show that the Punzi-net outperforms standard multivariate analysis techniques and generalises well to mass hypotheses for which it was not trained. This is achieved by training a single classifier that provides a coherent and optimal classification of all signal hypotheses over the whole search space. Our result constitutes a complementary approach to fully differentiable analyses in particle physics. We implemented this work using PyTorch and provide users full access to a public repository containing all the codes and a training example.