The U.S. Department of Energy (DOE) announced $16 million for 15 projects that will implement artificial intelligence (AI) methods to accelerate scientific discovery in nuclear physics research. DOE’s Argonne National Laboratory will lead two of those projects and be a collaborator on another.
These projects will use AI and Machine Learning (ML) tools and methods for nuclear physics experiments, simulation, theory and accelerator operation to expand and accelerate scientific reach.
“Artificial intelligence has the potential to shorten the timeline for experimental discovery in nuclear physics,” said Timothy Hallman, DOE associate director for the Office of Science for Nuclear Physics. “Particle accelerator facilities and nuclear physics instrumentation face a variety of technical challenges in simulations, control, data acquisition and analysis that artificial intelligence holds promise to address.”
The 15 projects will be conducted by nuclear physics researchers at eight DOE national laboratories and 22 universities. Projects will include the development of deep learning algorithms to identify a unique signal for studying physics of fundamental symmetry in extremely rare nuclear decays that if observed would demonstrate how our universe could have become dominated by matter rather than antimatter. Supported efforts also include AI-driven detector design for the Electron-Ion Collider (EIC) accelerator project under construction at DOE’s Brookhaven National Laboratory (BNL) that will probe the internal structure and forces of protons and neutrons that compose the atomic nucleus. Also, several accelerator beam optimization projects using AI/ML tools will be funded at scientific user facilities supported by Nuclear Physics including the Facility for Rare Isotope Beams at Michigan State University, the Relativistic Heavy Ion Collider at BNL, and the future EIC, to be located at BNL.
“Artificial intelligence has the potential to shorten the timeline for experimental discovery in nuclear physics,” — Timothy Hallman, DOE associate director for the Office of Science for Nuclear Physics
Argonne projects are:
- “Modern Data Analytics for the Large Gamma-Ray Spectrometers: GRETINA/GRETA and Gammasphere via ML and Optimization” (led by Argonne’s Mike Carpenter and the Argonne Tandem Linac Accelerator System (ATLAS) team)
- “Use of Artificial Intelligence to Optimize Accelerator Operations and Improve Machine Performance” (led by Argonne’s Brahim Mustapha)
- “STREAMLINE Collaboration: Machine Learning for Nuclear Many-Body Systems” (Led by Michigan State University with Argonne collaborator Alessandro Lovato)
The projects are supported by the DOE Office of Science Nuclear Physics Program. ATLAS is a DOE Office of Science user facility.
Awards were selected by competitive peer review. Total planned funding is $16 million, with $8 million in Fiscal Year 2023 dollars and outyear funding contingent on congressional appropriations. The list of other projects and more information can be found on the Nuclear Physics homepage.
