arXiv subjects

Garvita Agarwal

Publications and source records attributed to Garvita Agarwal.

6 recordsLinked to original sources

arXiv2026

Building an AI-native Research Ecosystem for Experimental Particle Physics: A Community Vision

Experimental particle physics seeks to understand the universe by probing its fundamental particles and forces and exploring how they govern the large-scale processes that shape cosmic evolution. This whitepaper presents a vision for how Artificial Intelligence (AI) can accelerate discovery in this field. We outline grand challenges that must be addressed to enable transformative breakthroughs and describe how current and planned experimental facilities can implement this vision to advance our understanding of the vast and complex physical world from the smallest to the largest scales. We show how facilities currently under construction, such as the HL-LHC, DUNE and soon EIC, can both benefit from and serve as proving grounds for this vision, while also enabling a longer-term goal for how future experiments -- like FCC-ee at CERN, IceCube-Gen2, a Muon Collider in the U.S., and smaller to mid-scale projects -- can be fully AI-native. We describe how a truly national-scale collaboration, jointly managed across large funding partners, and involving both DOE laboratories and universities, can make this happen.

hep-ex↗

arXiv2023

Jet Energy Scale and Resolution Measurements in CMS

Measurements of jet energy scale (JES) and resolution (JER) are presented, based on the legacy reconstruction of 13 TeV proton-proton collision data collected by the CMS experiment during the LHC Run 2 period from 2016-2018. Precision measurement of JES is of the utmost importance for the vast majority of physics measurements and searches at CMS. The high pileup, a harsh radiation environment, and time-dependent variations in detector response and calibration, all make precision JES measurement a challenging task. We present in-situ derivations of JES and JER based on CMS Run 2 data, as well as on simulated samples using various advanced techniques.

hep-ex↗

arXiv2022

Snowmass Early Career

The Snowmass 2021 strategic planning process provided an essential opportunity for the United States high energy physics and astroparticle (HEPA) community to come together and discuss upcoming physics goals and experiments. As this forward-looking perspective on the field often reaches far enough into the future to surpass the timescale of a single career, consideration of the next generation of physicists is crucial. The 2021 Snowmass Early Career (SEC) organization aimed to unite this group, with the purpose of both educating the newest generation of physicists while informing the senior generation of their interests and opinions. SEC is the latest in a series of the previously dubbed "Snowmass Young" organizations, from 2013 and 2001. This iteration has expanded on these efforts to significantly increase involvement and broaden the representation of the early career community in the process. Early career physicists are the future of the field. They will design, build, and operate next-generation experiments, and put in the work to usher in new discoveries. They are also disproportionately involved in work to improve the climate within HEPA. This document summarizes the work of SEC in consolidating a huge variety of physics perspectives and community opinions towards a bright, strategic future.

hep-ex↗

arXiv2022

Environment Dependent Electron Mass and the H$_0$-Tension

A difference in the electron mass at recombination from the measured value today has been shown to relieve the Hubble tension. In this article we propose a simple mechanism that would allow for such a mass difference. Similar in form to the Higgs field, we consider a Yukawa coupling between the symmetron scalar field and the electron. However, the symmetron vacuum expectation value depends on the matter density of its environment which, in principle, would allow its contribution towards the electron mass to differ between recombination and today. We detail the coupling and discuss possible observables.

hep-ph↗

arXiv2022

Snowmass 2021 Community Survey Report

The Snowmass Community Survey was designed by the Snowmass Early Career (SEC) Survey Core Initiative team between April 2020 and June 2021, and released to the community on June 28, 2021. It aims to be a comprehensive assessment of the state of the high-energy particle and astrophysics (HEPA) community, if not the field, though the Snowmass process is largely based within the United States. Among other topics, some of the central foci of the Survey were to gather demographic, career, physics outlook, and workplace culture data on a large segment of the Snowmass community. With nearly $1500$ total interactions with the Survey, the SEC Survey team hopes the findings and discussions within this report will be of service to the community over the next decade. Some conclusions should reinforce the aspects of HEPA which are already functional and productive, while others should strengthen arguments for cultural and policy changes within the field.

hep-ex↗

arXiv2021

Explainable AI for ML jet taggers using expert variables and layerwise relevance propagation

A framework is presented to extract and understand decision-making information from a deep neural network (DNN) classifier of jet substructure tagging techniques. The general method studied is to provide expert variables that augment inputs ("eXpert AUGmented" variables, or XAUG variables), then apply layerwise relevance propagation (LRP) to networks both with and without XAUG variables. The XAUG variables are concatenated with the intermediate layers after network-specific operations (such as convolution or recurrence), and used in the final layers of the network. The results of comparing networks with and without the addition of XAUG variables show that XAUG variables can be used to interpret classifier behavior, increase discrimination ability when combined with low-level features, and in some cases capture the behavior of the classifier completely. The LRP technique can be used to find relevant information the network is using, and when combined with the XAUG variables, can be used to rank features, allowing one to find a reduced set of features that capture part of the network performance. In the studies presented, adding XAUG variables to low-level DNNs increased the efficiency of classifiers by as much as 30-40\%. In addition to performance improvements, an approach to quantify numerical uncertainties in the training of these DNNs is presented.

physics.data-an↗