Deep Learning Methods for Fundamental Physics
Benjamin Nachman, Research Fellow
Berkeley Institute for Data Science (BIDS)
Applications for Fall 2024 are closed for this project.
This is an exciting time in fundamental physics: there are many experimental and theoretical hints for new phenomena (such as dark matter), yet we do not yet have any significant evidence for new particles or forces of nature since the discovery of the Higgs Boson in 2012. This could be because our experiments are not sensitive enough, that the new particles are rare, or that we are not looking in the right place. The goal of this project is to develop, adapt, and deploy state-of-the-art deep learning methods to enhance the search for new particles and/or to explore emergent properties of fundamental interactions.
Our group has developed a variety of deep learning methods to automatically explore high-dimensional physics data. This URAP project will involve extending and/or applying these techniques to a variety of physical systems including collider physics (proton-proton, electron-proton, and electron-positron colliders such as the Large Hadron Collider, the Electron Ion Collider, and the International Linear Collider), neutrino physics, and astroparticle physics (e.g. the Gaia space observatory).
Role: The exact work will depend on the experience, availability, interest, and progress of the student. Research in this area is at the intersection of theory, experiment, and applied statistics/machine learning. At least 6-8 hours are typically needed to make significant progress.
Qualifications: Applicants should be interested in particle-, nuclear-, and/or astrophysics and machine learning solutions to physics challenges. Majoring/minoring in Physics or Astronomy would be great, but this is not required; majors in EECS/CS/Data Science/Math/Statistics or related disciplines would be most welcome with significant interest in physics topics. Experience with at least one programming language (Java/C++/Matlab/Python/Julia/etc.) is required. The research will mostly be carried out in Python, so this would be desired but is not required. Great teamwork (e.g. communication skills, punctuality, organization) are necessary for a success. While not required, recommended skills include familiarity with basic probability and statistics, experience with communication and collaboration tools like Slack and Github, experience with deep learning packages like Keras/Tensorflow or PyTorch. If you have any experience with these topics, please mention it in your application.
Hours: to be negotiated
Off-Campus Research Site: Our group sits in Building 50 at Berkeley Lab (20 min walk / 10 min shuttle ride up Hearst from the Physics Department). In-person is an option as is virtual. We make heavy use of Slack.
Related website: http://bpnachman.com
Related website: http://www.physics.lbl.gov/machinelearning/