neos: End-to-End-Optimised Summary Statistics for High Energy Physics

TL;DR

This paper introduces neos, a fully differentiable high-energy physics workflow that optimizes summary statistics end-to-end, improving analysis sensitivity while accounting for systematic uncertainties.

Contribution

The work presents neos, a novel differentiable framework for optimizing summary statistics in high-energy physics analyses, integrating systematic uncertainty treatment.

Findings

Optimizes summary statistics for better analysis sensitivity

Incorporates systematic uncertainties into the optimization process

Demonstrates end-to-end differentiability in physics workflows

Abstract

The advent of deep learning has yielded powerful tools to automatically compute gradients of computations. This is because training a neural network equates to iteratively updating its parameters using gradient descent to find the minimum of a loss function. Deep learning is then a subset of a broader paradigm; a workflow with free parameters that is end-to-end optimisable, provided one can keep track of the gradients all the way through. This work introduces neos: an example implementation following this paradigm of a fully differentiable high-energy physics workflow, capable of optimising a learnable summary statistic with respect to the expected sensitivity of an analysis. Doing this results in an optimisation process that is aware of the modelling and treatment of systematic uncertainties.