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

arXiv: https://arxiv.org/abs/2203.05570 · HTML: https://mlsanity.com/paper/2203.05570/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 2 of its 8 references that are in the map.
- First posted on arXiv 10 Mar 2022.

## Builds on (most cited first, 2 of 2)

- [Learning to Pivot with Adversarial Networks](https://mlsanity.com/paper/1611.01046/) (2016)
- [INFERNO: Inference-Aware Neural Optimisation](https://mlsanity.com/paper/1806.04743/) (2018)

## Built on by (most cited first, 10 of 10)

- [Finetuning Foundation Models for Joint Analysis Optimization](https://mlsanity.com/paper/2401.13536/) (2024)
- [Branches of a Tree: Taking Derivatives of Programs with Discrete and Branching Randomness in High Energy Physics](https://mlsanity.com/paper/2308.16680/) (2023)
- [Bias and Priors in Machine Learning Calibrations for High Energy Physics](https://mlsanity.com/paper/2205.05084/) (2022)
- [FAIR Universe HiggsML Uncertainty Dataset and Competition](https://mlsanity.com/paper/2410.02867/) (2024)
- [Supercharging Simulation-Based Inference for Bayesian Optimal Experimental Design](https://mlsanity.com/paper/2602.06900/) (2026)
- [Unknown Unknowns: Model Misspecification in Machine Learning for Physics](https://mlsanity.com/paper/2608.13633/) (2026)
- [Exploring the Universality of Hadronic Jet Classification](https://mlsanity.com/paper/2204.03812/) (2022)
- [Designing Observables for Measurements with Deep Learning](https://mlsanity.com/paper/2310.08717/) (2023)
- [Learning to bin: differentiable and Bayesian optimization for multi-dimensional discriminants in high-energy physics](https://mlsanity.com/paper/2601.07756/) (2026)
- [Hybrid Neural Simulation-Based Inference for Robust Applications and Limited-Budget Scenarios](https://mlsanity.com/paper/2609.36044/) (2026)

## Abstract (arXiv)

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.