Project Grant 2610087

The National Science Foundation Directorate for Mathematical and Physical Sciences awarded the University of Hawaii $300,000 on August 1, 2026, under the Mathematical and Physical Sciences program (CFDA 47.049) to develop real-time artificial intelligence and machine-learning algorithms for the KamLAND2-Zen neutrinoless double-beta decay detector experiment.

The University of Hawaii group will deploy AI/ML inference directly on radio-frequency system-on-chip and multi-processor system-on-chip devices to enable the KamLAND2-Zen detector to identify physics events more efficiently while rejecting background signals. The algorithms will extract photon hit times and charge across all photomultiplier tube channels at the trigger level, allowing intelligent, fast decisions from large detector data streams in real time. The same hardware–AI codesign methods will have broad applications across particle and nuclear experiments requiring rapid signal processing. The project will also train a new generation of data physicists combining deep physics knowledge with modern AI skills, with the laboratory system serving as a real-time AI platform for student research and education in an EPSCOR jurisdiction.

KamLAND2-Zen, scheduled to begin data taking in 2027, searches for neutrinoless double-beta decay of xenon-136 using a xenon-loaded liquid scintillator detector. The physics objective is to determine whether the neutrino is its own antiparticle, providing insight into neutrino mass origin and the cosmic matter-antimatter asymmetry.

Place of performance is Honolulu, Hawaii. The period of performance runs from August 1, 2026, through July 31, 2029.