High Energy Physics Seminar
This seminar will be held via Zoom.
> Meeting ID: 381 132 649
Three years ago, standard model production of the Z (or W) boson decaying to light quarks had not been observed at the Large Hadron Collider; it was thought to be impossible. We present a new technique to observe these productions and we observe a clear W and Z peak and, with the addition of Deep Learning, present new ideas to perform foundational measurements of the Higgs boson and Dark Matter. From these results, it becomes clear that the LHC is severely limiting our physics output; we present a new approach to overcome these limitations through the use of a heterogeneous computing model relying on FPGAs, GPUs, and CPUs. Beyond the LHC, we look at the broader scope of accelerated computing in physics covering LIGO, Neutrino physics, and the LHC and discuss how deep learning can lead the way to a fundamental change in how we process our data. Finally, we apply deep learning to resonances decaying to b-quarks and we present a new and intriguing result on gluon fusion Higgs production to b-quarks with the full Run 2 LHC dataset.