ALPhA @ Davidson College
Welcome to the Algorithms for Learning in Physics Applications (ALPhA) research group at Davidson College, led by Dr. Michelle Kuchera and Dr. Raghu Ramanujan. We develop deep learning methods for nuclear physics and particle physics, working closely with experimentalists at the Facility for Rare Isotope Beams (FRIB) and other national laboratories and universities. Undergraduate researchers play a central role in our work.
What we're working on
Representation Learning for TPC Data
Reusable representations of time projection chamber data. Our recent work examines how model architecture and training shape these embeddings across tasks, experiments, and detectors, including the AT-TPC and GADGET II.
Bridging Simulation and Experiment
Generative and diffusion models that translate between simulated and measured detector data, adding realistic detector response to simulations or removing noise from measurements.
Finding Rare Events in the AT-TPC
A point-cloud model trained on heuristically labeled data that identifies rare fission events a hand-tuned method left unclassified. Together, the two approaches identify about 99% of fission events.
Earlier ALPhA projects include machine-learning event generators and QCD analysis with Jefferson Lab and jet emulation for the CMS experiment at CERN. See our publications for more.
Recent News
Interested in ALPhA?
Are you a student interested in working with us? We welcome Davidson students from all areas of study who are curious about the intersection of AI and science. There are potential opportunities for non-Davidson undergraduates. Reach out to Dr. Michelle Kuchera or Dr. Raghu Ramanujan to learn more.
Are you a physics researcher looking for potential collaboration? We are enthusiastic about collaborating on projects that we find exciting from a machine learning perspective. Please read Collaborating with ALPhA, then reach out to Michelle or Raghu to discuss your project.
Contact: Dr. Michelle Kuchera · Dr. Raghu Ramanujan
Funding
We gratefully acknowledge support for ALPhA's research from:
- National Science Foundation, Award 2311263: Elements: Portable Machine Learning Models for Experimental Nuclear Physics (2023–2027)
- National Science Foundation, Award 2012865: RUI: Machine Learning Approaches for Accelerating Scientific Discovery in Nuclear Physics (2020–2024)
- U.S. Department of Energy, Laboratory Directed Research and Development, through Thomas Jefferson National Accelerator Facility: Constructing Prototypes for a Universal Monte Carlo Event Generator (2019–2021)
- Center for Nuclear Femtography, through Old Dominion University: QCD Theory and Machine Learning for Global Analysis (2019–2021)
- Facility for Rare Isotope Beams: summer research support (2017) and Visiting Scholar Award (2019)
- Davidson College: Davidson Research Initiative, Faculty Study & Research grants, and the Research in Science Experience (RISE) program