Dr. Nhan Tran, Fermilab, US

Personal page: https://nhanvtran.wixsite.com/home

We will discuss the growing research area at the intersection of machine learning, detector hardware, and heterogeneous computing. We envision this will advance the overall output and sensitivity of particle physics experiments, enable intelligent and autonomous detectors, and solve mounting online and offline computing challenges. Two specific examples will be given. First, we will discuss neural networks implemented in on-detector electronics for powerful real-time filtering at the LHC. Then, we will discuss a recent result for GPU-accelerated neutrino computing workflows accelerating ProtoDUNE reconstruction by nearly 3x with optimized GPU resource utilization.