Projects

Research Projects

HSF-India Trainee Project

Fig: Artistic rendition of an FCC event

Project: Building examples for Future Circular Collider (FCC) analyses using the Columnar Framework For Effective Analysis (COFFEA) framework and developing the schema class implementation of FCC simulation samples in COFFEA
Mentors: David Lange (Princeton University), Bhawna Gomber (University of Hyderabad)
Duration: June 2023 - Ongoing
Description: :
The purpose of this project is to build the necessary schema class for FCC simulation samples in COFFEA and create examples of simple analyses done in COFFEA utilizing FCC simulation samples. The schema classes for the NanoAOD data structure are already present in COFFEA. Developing a schema class for FCC samples would enable analyses to be performed in Python and take advantage of parallel processing with less time-to-insight. Based on the existing examples of FCC analyses, COFFEA examples of the same are to be built and documented, which will streamline the reproducibility of the results.
Links:

1. https://research-software-collaborations.org/trainees/PrayagYadav.html (opens in a new tab)
2. https://prayagyadav.github.io/coffea-fcc-analyses/ (opens in a new tab)
3. https://github.com/prayagyadav/coffea-fcc-analyses (opens in a new tab)

Master's Thesis

Fig: CMS Event Display

Project Title: “Search for dark matter in the Higgs to two bottom jets channel with large missing transverse energy.”
Supervisor: Dr Bhawna Gomber, University of Hyderabad
Duration: July 2023 to May 2024
Description:

• Among the many open questions in physics, the question of what Dark Matter is, is a mysterious one.
• In this thesis, I study a specific 'simplified model' of dark matter. The simplified model is a supersymmetric model called the two Higgs doublet model - a (2HDMa). The 2HDMa model introduces a new pseudoscalar 'a' which mediates the interaction between the Dark sector (denoted by $\chi$ and $\bar{\chi}$) , and the usual standard model particles.
• The signal signature is a high missing transverse momentum and a standard model Higgs which decays to two bottom quarks. These bottom quarks produce particle jets which are identified as 'b-jets' using a b- tagging algorithm.
• The analysis involves the identification of the signal (large $p_T^{miss} + H \rightarrow b\bar{b}$ ) from various backgrounds which are estimated by defining control regions and by Monte-Carlo simulation.
• I use the COFFEA framework (Columnar Object Framework For Effective Analysis) as my analysis tool.
• The analysis is originally going on at the University of Wisconsin, Madison. I collaborate with the Wisconsin's group to perform control region studies.
• The two major backgrounds for this analysis are top pairs ($t\bar{t}$) and $Z \rightarrow \nu\bar{\nu}$ . These two backgrounds are estimated by estimating four different control regions: single muon, single electron, double muon and double electron. Currently I am working on estimating the muon control region.
• My aim is to push the analysis forward by helping with control region estimation.

Internship with CMS group at University of Hyderabad

Fig: The Compact Muon Solenoid design

Supervisor: Dr Bhawna Gomber
Duration: 20th July 2022 to 4th July 2023

• Before starting my masters project, I did an internship with the CMS (Compact Muon Solenoid) Group at the University of Hyderabad. I did this in parallel to my coursework classes.
• During this time, I learnt the basics of analysis in experimental HEP.
• I learnt a lot of tools: C++, ROOT, CMSSW, Python, COFFEA, bash scripting and submitting jobs to dask and HTCondor schedulers.
• I took up the job of setting up and maintaining the newly bought server for my lab. I learnt how to install RHEL based enterprise Linux on the server, installed ROOT and other HEP relevant software, deployed containerized solutions to HEP analysis like singularity and docker containers, hosted an internal Jupyterhub server with containerized backends.
• At the physics analysis part of things, I helped the PhD students with estimation of dielectron backgrounds in their analyses using ROOT and C++.

Project in experimental nuclear physics

Fig: Compton scattering angles

Title: 'Study of entanglement of photons in para-positronium decay and its implications'
Supervisor: Prof. Rudrajyoti Palit
Duration: (May 2021 to July 2021) and (Dec 2021 to Jan 2022)

• I did a reading project on the above-mentioned title. I learnt in detail about Gamma-ray spectroscopy and double Compton scattering cross sections.
• I focused on the use of segmented HPGe detectors to detect the correlation between two entangled photons produced as a result of a positronium decay ($e^+ + e^-$).
• I learnt about the signal processing and data acquisition required for such an analysis. I also learnt about CZT detectors and $NaI(Tl)$ gamma spectroscopy.
• At the end of my project, my guide posed a problem. The problem was to find the relative contributions of Compton Scattering and Photoelectric absorption in a full photopeak corresponding to an $E=0.511MeV$. I solved the problem by creating a simulation in python from scratch.

Personal Projects

Hypothesis Testing, Linear Regression and Multiple Regression

Currently I am learning statistical learning from the Robert Tibshirani's course : Introduction to Statistical learning with python. As a part of that, I will gather inferences and insights from a dataset obtained from kaggle using linear regression. Checkout my progress at this repository (opens in a new tab).