Publications
Publication Information
| Title | Is infrared-collinear safe information all you need for jet classification? |
| Authors | Dimitrios Athanasakos, Andrew Larkoski, James Mulligan, Mateusz Ploskon, Felix Ringer |
| JLAB number | JLAB-THY-23-3818 |
| LANL number | (None) |
| Other number | DOE/OR/23177-6148 |
| Document Type(s) | (Journal Article) |
| Associated with EIC: | Yes |
| Supported by Jefferson Lab LDRD Funding: | No |
| Funding Source: | Nuclear Physics (NP) |
|
Journal Compiled for Journal of High Energy Physics | |
| Publication Abstract: | Machine learning-based jet classifiers are able to achieve impressive tagging performance in a variety of applications in high energy and nuclear physics. However, it remains unclear in many cases which aspects of jets give rise to this discriminating power, and whether jet observables that are calculable in perturbative QCD such as those obeying infrared-collinear (IRC) safety serve as sufficient inputs. In this article, we introduce a new classifier, Jet Flow Networks (JFNs), in an effort to address the question of whether IRC unsafe information provides additional discriminating power in jet classification. JFNs are permutation-invariant neural networks (deep sets) that take as input the kinematic information of reconstructed subjets. The subjet radius serves as a tunable hyperparameter, enabling the sensitivity to soft emissions and nonperturbative effects to be gradually increased as the subjet radius is decreased. We demonstrate the performance of JFNs for quark vs. gluon and QCD vs. $Z$ jet tagging. For small subjet radius, the performance of JFNs is comparable to the IRC-unsafe Particle Flow Networks (PFNs), demonstrating that infrared-collinear unsafe information is not necessary to achieve strong discrimination. As the subjet radius is increased, the performance of the JFNs remains essentially unchanged until physical thresholds that we identify are crossed. For relatively large subjet radii, we show that the JFNs may offer an increased model independence with a modest tradeoff in performance compared to classifiers that use the full particle information of the jet. Our results shed new light onto how machines learn patterns in high-energy physics data. |
| Experiment Numbers: | |
| Group: | THEORY CENTER |
| Document: | |
| DOI: | |
| Accepted Manuscript: | |
| Supporting Documents: | |
| Supporting Datasets: | |