Haichen Wang, Professor

Closed (1) Search for New Physics via Analysis of Data Collected by the ATLAS Experiment at the LHC

Applications for fall 2021 are now closed for this project.

The Large Hadron Collider (LHC) is the most powerful particle accelerator ever built and researchers use its data to study what the universe was like shortly after the big bang. Researchers at Berkeley and the Lawrence Berkeley National Laboratory (LBNL) play a key role in all aspects of the ATLAS experiment, one of two main detectors at the LHC. Students who join our group work on a variety of projects, ranging from detector development to data analysis.

There are about a dozen ATLAS researchers stationed in Berkeley at any one time and there are a variety of projects available. Picking one of the general projects listed below does not commit you to a particular researcher or project. Projects and mentors are decided during and after the interview process.

This program is expected to span over multiple semesters. The first semester will be a training program to bring student researchers up to speed. The focus will be on understanding basic concepts, learning software tools such as ROOT, and developing good research practice and work ethic. Once students are trained, they would be embedded in research projects to work with postdocs and graduate students, with the goal of completing an ATLAS data analysis project.

Qualifications: Students are expected to be self-motivated and should be good at time management. Students with heavy course load during the semester is discouraged. C++ and Python are used extensively in our data analysis. Please indicate your level and experience with programming languages in the application Prior experience with machine learning algorithms and/or applications is desirable but not required. Please indicate in your application if you have any relevant experience. Knowledge about particle physics (Physics-129) is desirable but not required. Applicants should fill in this survey: https://forms.gle/a9HszXz4jAocEgC79 Ideal applicants should be able to commit 12 hours or more weekly to the project, however, exceptions may be made on a case by case basis.

Weekly Hours: to be negotiated

Related website: http://hwang43.web.cern.ch/hwang43/
Related website: https://atlas.cern/discover/physics

Closed (2) Development and validation of advanced numerical models for superconducting magnets

Applications for fall 2021 are now closed for this project.

Do you want to contribute to the development of future particle colliders?

We are currently developing key analysis tools for future superconducting magnets.

The main scope of these tools will be the ability to predict the superconting coil mechanical properties and failure mechanisms during their operation.

We are looking for highly motivated students to participate in the development and experimental validation of these tools. Ideally, the student would have the following skills:

A good understanding of structural mechanics fundamentals
Basic coding/data processing in Matlab and/or Python

The activities will be performed partially remotely, and might include some or all of the following:

- Numerical/analytical modeling of superconducting coils
- Design and construction of experimental tests to validate the models
- Data acquisition and analysis of the measurements

Weekly Hours: to be negotiated

Off-Campus Research Site: This project will be performed remotely and will be directed by LBNL scientist Giorgio Vallone, et al.

Closed (3) Creating a picture of the Higgs Boson

Applications for fall 2021 are now closed for this project.

The Higgs boson is a unique elementary particle responsible for creating masses for other elementary particles in our Universe. In 2012, the Higgs boson was discovered at the Large Hadron Collider (LHC) by the ATLAS and CMS experiments, which led to the 2013 Nobel Prize in Physics being awarded to Peter Higgs and Francois Englert, two theorists who predicted the existence of this particle in 1964. Since then, the experimental study of the Higgs boson has been the frontier of our exploration of elementary particles and fundamental interactions.

This project invites undergraduate students who are proficient in Python and/or C++ programming, have experience computer graphics and data visualization, and have interests in particle physics, to work on the generation of the very first "picture" of the Higgs boson using data collected by the ATLAS experiment during the Run-2 of the LHC.

The Higgs boson produced at the LHC decays to other elementary particles. Among all the possibilities, the Higgs boson can decay to a pair of photons, which are well measured experimentally. During the LHC Run-2, a period from 2015 to 2018, approximately 16,000 Higgs bosons were produced and decayed to photon pairs. These photons are of extremely high energy, far beyond that of visible lights. We will use these Higgs boson to photons decays to create an image of the Higgs boson. This would be the first picture of the Higgs boson using real collision data.

process analyze collision data from the LHC
create 2-D and 3-D visualization of Higgs boson data

Qualifications: proficient in Python or C++ (Physics 77 or equivalent) experience with data visualization and/or computer graphics strong interests in physics. knowledge in particle physics desirable but not required Applicants should provide a link to their github repository where a previous sample project is stored. We would like to get a sense of applicant's coding experience

Weekly Hours: to be negotiated

Off-Campus Research Site: This project can be done remotely during Shelter-in-Place.

Related website: http://hwang43.web.cern.ch/hwang43/
Related website: https://atlas.cern/discover/physics

Closed (4) Generator tuning with Uncertainties

Applications for fall 2021 are now closed for this project.

Monte Carlo (MC) event generators are essential tools for analyzing collision events at the Large Hadron Collider. In these generators, there are a number of relatively free parameters, collectively called generator parameters, which must be tuned if the generator is to describe experimental data. Our group has developed a tool in Python that can automatically tune these generator parameters. However, the current tuning procedure misses a key component by construction, the uncertainties associated with MC generators. The project is to integrate the MC uncertainties into the tuning procedure and estimate their impact on the generator parameters.

This work involves generating simulated events with different MC generator parameters, analyzing these events with pre-defined analyses, tuning the MC generator parameters, and discussing the results with researchers at LBNL. Actively learning about MC generators and parallel computing is an important part of this work; related materials will be made available.

Qualifications: Applicants should be interested in particle physics. Applicants should finish classes related to Probabilities and Statistics. Knowing sensitive analysis and uncertainty estimation are strong pluses. Majoring/Minoring in Physics would be great. The research will mostly be carried out in Python, sometimes in C++, so familiarity with both is desired.

Weekly Hours: 9-11 hrs

Off-Campus Research Site: Our default mode of operation will be virtual, meeting on Zoom and communicating via Slack and email. This research project will be directed by Dr. Xiangyang Ju of Lawrence Berkeley National Lab.