# Adversarially-trained autoencoders for robust unsupervised new physics searches

## TL;DR

This paper introduces an adversarially-trained autoencoder approach to improve the robustness of unsupervised anomaly detection in particle physics, specifically targeting resonance-induced t-tbar final states, by reducing sensitivity to experimental uncertainties.

## Contribution

The paper proposes combining autoencoders with adversarial neural networks to mitigate the impact of experimental uncertainties on anomaly detection in particle physics searches.

## Key findings

- Achieves robust anomaly detection in resonance-induced t-tbar states.
- Reduces sensitivity of autoencoders to experimental smearing.
- Demonstrates effectiveness of adversarial training in particle physics anomaly detection.

## Abstract

Machine learning techniques in particle physics are most powerful when they are trained directly on data, to avoid sensitivity to theoretical uncertainties or an underlying bias on the expected signal. To be able to train on data in searches for new physics, anomaly detection methods are imperative, which can be realised by an autoencoder acting as an unsupervised classifier. The last source of uncertainties affecting the classifier are then experimental uncertainties in the reconstruction of the final-state objects. To mitigate their effect on the classifier and to allow for a realistic assessment of the method, we propose to combine the autoencoder with an adversarial neural network to remove its sensitivity to the smearing of the final-state objects. We quantify its effect and show that one can achieve a robust anomaly detection in resonance-induced $t\bar{t}$ final states.