arXiv subjects

Francesco Rubbo

Publications and source records attributed to Francesco Rubbo.

3 recordsLinked to original sources

arXiv2025

Neural Posterior Unfolding

Differential cross section measurements are the currency of scientific exchange in particle and nuclear physics. A key challenge for these analyses is the correction for detector distortions, known as deconvolution or unfolding. Binned unfolding of cross section measurements traditionally rely on the regularized inversion of the response matrix that represents the detector response, mapping pre-detector (`particle level') observables to post-detector (`detector level') observables. In this paper we introduce Neural Posterior Unfolding, a modern, Bayesian approach that leverages normalizing flows for unfolding. By using normalizing flows for neural posterior estimation, NPU offers several key advantages including implicit regularization through the neural network architecture, fast amortized inference that eliminates the need for repeated retraining, and direct access to the full uncertainty in the unfolded result. In addition to introducing NPU, we implement a classical Bayesian unfolding method called Fully Bayesian Unfolding (FBU) in modern Python so it can also be studied. These tools are validated on simple Gaussian examples and then tested on simulated jet substructure examples from the Large Hadron Collider (LHC). We find that the Bayesian methods are effective and worth additional development to be analysis ready for cross section measurements at the LHC and beyond.

hep-ph↗

arXiv2018

A Bottom Line for the LHC Data by Leveraging Pileup as a Zero Bias Trigger

Due to a limited bandwidth and a large proton-proton interaction cross-section relative to the rate of interesting physics processes, most events produced at the Large Hadron Collider (LHC) are discarded in real time. A sophisticated trigger system must quickly decide which events should be kept and is very efficient for a broad range of processes. However, there are many processes that cannot be accommodated by this trigger system. Furthermore, there may be models of physics beyond the Standard Model (BSM) constructed after data taking that could have been triggered, but no trigger was implemented at run time. Both of these cases can be covered by exploiting pileup interactions as an effective zero bias sample. At the end of High-Luminosity LHC operations, this zero bias dataset will have accumulated about 1/fb of data from which a bottom line cross-section limit of O(1) fb can be set for BSM models already in the literature and those yet to come.

hep-ph↗

arXiv2017

Weakly Supervised Classification in High Energy Physics

As machine learning algorithms become increasingly sophisticated to exploit subtle features of the data, they often become more dependent on simulations. This paper presents a new approach called weakly supervised classification in which class proportions are the only input into the machine learning algorithm. Using one of the most challenging binary classification tasks in high energy physics - quark versus gluon tagging - we show that weakly supervised classification can match the performance of fully supervised algorithms. Furthermore, by design, the new algorithm is insensitive to any mis-modeling of discriminating features in the data by the simulation. Weakly supervised classification is a general procedure that can be applied to a wide variety of learning problems to boost performance and robustness when detailed simulations are not reliable or not available.

hep-ph↗