An official website of the United States government
Many machine learning applications involve learning a latent representation of data, which is often high-dimensional and difficult to directly interpret. In this work, we propose “moment pooling,” a natural extension of deep sets networks which drastically decreases the latent space dimensionality of these networks while maintaining or even improving performance. Moment pooling generalizes the summation in deep sets to arbitrary multivariate moments, which enables the model to achieve a much higher effective latent dimensionality for a fixed learned latent space dimension. We demonstrate moment pooling on the collider physics task of quark/gluon jet classification by extending energy flow networks (EFNs) to moment EFNs. We find that moment EFNs with latent dimensions as small as 1 perform similarly to ordinary EFNs with higher latent dimension. This small latent dimension allows for the internal representation to be directly visualized and interpreted, which in turn enables the learned internal jet representation to be extracted in closed form. Published by the American Physical Society2024
more »
« less
Moment extraction using an unfolding protocol without binning
Deconvolving (“unfolding”) detector distortions is a critical step in the comparison of cross-section measurements with theoretical predictions in particle and nuclear physics. However, most existing approaches require histogram binning while many theoretical predictions are at the level of statistical moments. We develop a new approach to directly unfold distribution moments as a function of another observable without having to first discretize the data. Our moment unfolding technique uses machine learning and is inspired by Boltzmann weight factors and generative adversarial networks (GANs). We demonstrate the performance of this approach using jet substructure measurements in collider physics. With this illustrative example, we find that our moment unfolding protocol is more precise than bin-based approaches and is as or more precise than completely unbinned methods. Published by the American Physical Society2024
more »
« less
- Award ID(s):
- 2019786
- PAR ID:
- 10570057
- Publisher / Repository:
- American Physical Society
- Date Published:
- Journal Name:
- Physical Review D
- Volume:
- 110
- Issue:
- 11
- ISSN:
- 2470-0010
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
A computationally efficient method for calculating the transport of neutrino flavor in simulations is to use angular moments of the neutrino one-body reduced density matrix, i.e., “quantum moments.” As with any moment-based radiation transport method, a closure is needed if the infinite tower of moment evolution equations is truncated. We derive a general parametrization of a quantum closure and the limits the parameters must satisfy in order for the closure to be physical. We then derive from multiangle calculations the evolution of the closure parameters in two test cases which we then progressively insert into a moment evolution code and show how the parameters affect the moment results until the full multiangle results are reproduced. This parametrization paves the way to setting prescriptions for genuine quantum closures adapted to neutrino transport in a range of situations. Published by the American Physical Society2025more » « less
-
Physics Education Research has a rich history of identifying common student ideas about specific physics topics. In the context of electric circuits, existing research on students’ ideas has primarily focused on misconceptions, misunderstandings, and difficulties. In this paper, we take a resource-oriented approach to identifying common student ideas about circuits by characterizing ideas we see as generative “seeds of science” that could form the basis of more sophisticated understandings. Based on our analysis of 1557 university physics student responses to five conceptual questions, we identify four common resources for understanding circuits. Published by the American Physical Society2024more » « less
-
Network analysis has become a well-recognized methodology in physics education research (PER), with study topics including student performance and persistence, faculty change, and the structure of conceptual networks. The social network analysis side of this work has focused on quantitative analysis of whole-network cases, such as the structure of networks in single classrooms. Egocentric or personal network approaches are largely unexplored, and qualitative methods are underdeveloped. In this paper, we outline theoretical and practical differences between two major network paradigms—whole-network and egocentric—and introduce theoretical frameworks and methodological considerations for egocentric studies. We also describe qualitative and mixed-methods approaches that are currently missing from the PER literature. We identify areas where these additional network methods may be of particular interest to physics education researchers and end by discussing example cases and implications for new PER studies. Published by the American Physical Society2024more » « less
-
The Forward Search Experiment (FASER) at CERN’s Large Hadron Collider (LHC) has recently directly detected the first collider neutrinos. Neutrinos play an important role in all FASER analyses, either as signal or background, and it is therefore essential to understand the neutrino event rates. In this study, we update previous simulations and present prescriptions for theoretical predictions of neutrino fluxes and cross sections, together with their associated uncertainties. With these results, we discuss the potential for possible measurements that could be carried out in the coming years with the FASER neutrino data to be collected in LHC Run 3 and Run 4. Published by the American Physical Society2024more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1103/physrevd.110.116013
