Illuminating the Dark Matter Search with Edge AI
Radiation and particle detectors let us observe the universe beyond what our senses can achieve. Scientists come up with new ways to collect more and more information about our universe using detectors to transform these particles into electrical signals and binary bits. Dark matter detectors are a particularly active venue for detector development, and the ARGO detector will exploit novel photon-to-digital converters to detect light emission in a liquid argon target. The large surface area (~200 m2) and the immense number of detectors needed to cover this surface at the µm2 scale requires the development of a new data acquisition paradigm.
Transmitting the immense amount of raw data continuously to a computing system would be prohibitive both in material and energy costs. My research group develops systems to distribute real-time compression and analysis algorithms along the data acquisition chain including right up at the edge of the detector. This approach presents several challenges with limited computing power and memory availability. We combine small AI models with carefully selected signal processing and compression algorithms to achieve very efficient smart acquisition systems capable of several orders of magnitude data reduction in a few microseconds.
Bio
Audrey Corbeil Therrien is an Assistant Professor in the Department of Electrical and Computer Engineering at the Université de Sherbrooke in Québec, Canada. She currently holds the Tier-2 Canada Research Chair in real-time embedded intelligence for ultra-high rate detectors. She has worked at CERN (2015) in Geneva, Switzerland and at the SLAC National Accelerator Laboratory where she held a Banting Fellowship (2018-2020). Her research aims to improve real-time analyses at the edge with the integration of machine learning in high performance radiation instrumentation systems. She is part of several conference committees, and she is co-chair for the 2023 IEEE Nuclear Science Symposium. She has been strongly involved with the promotion of technical and scientific careers for diverse groups with Université de Sherbrooke, SLAC, Stanford University, local non-profits and at several conferences. She received several awards, including the 2017 IEEE Glenn F. Knoll Graduate Educational Grant, the 2019 Best Thesis Award in Science and Engineering at the Université de Sherbrooke and a finalist for the 2023 Prix Honoris Genius — Relève of the Ordre des Ingénieurs du Québec.