2018

Starting in 2021, the Large Hadron Collider (LHC) will increase its operational energy, luminosity and number of interactions per collision. These conditions result in a factor of 10 more physics events, and up to 25% more overlapping processes in each event. This environment will make it difficult for the ATLAS trigger system to efficiently decide which data to store. This talk describes how a new hardware system, called the ATLAS Fast TracKer (FTK), will allow ATLAS to keep up with the demanding LHC conditions. Analyses that benefit from the FTK’s potential to improve b-quark and tau-lepton identification are discussed. These analyses have the potential to shed light on B-anomalies and gravity’s role in the Standard Model.

Readout upgrades that are aimed at improving electron and muon measurements are briefly discussed. Analyses that could benefit electron, muon, b-quark and tau identification include searches that address how neutrino masses and dark matter fit into the Standard Model.

Two outstanding questions in physics are the nature of dark matter and the origin of the matter- antimatter asymmetry in the universe. Astrophysical observations imply the existence of dark matter, an invisible and dominant mass component in the universe, but it has eluded direct detection to date. A measurement of charge-parity (CP) violation in the lepton sector may help explain the observed preponderance of matter over antimatter. I will present new results from the XENON1T dark matter search experiment, consisting of a multi-tonne dual-phase (liquid-gas) xenon time projection chamber, as well as my future plans for the measurement of CP violation by the T2K long-baseline neutrino oscillation experiment and the next-generation large water Cherenkov detector, Hyper-Kamiokande, attempting to answer these two questions.

I will survey the status of Deep Learning in High Energy Physics, with focus on new emerging techniques. I will discuss the path from the current state of mostly proof of principle and feasibility studies to physics results incorporating the new techniques and eventually incorporating Deep Learning into the production workflow of HEP experiments. And I will propose future Deep Learning projects in HEP.