Seminars and Colloquia
Physics
Searching for low mass/HT resonances at the LHC using unconventional approaches
Thu, Jan 23, 2020,
04:00 PM
at Seminar Hall 31, 2nd Floor, Main Building
Abhijith Gandrakota
Rutgers University
Many BSM models predict low mass particles whose decays result in multijet resonances. But high trigger thresholds on hadronic triggers can blind CMS to these resonances. CMS has implemented Data Scouting techniques to regain sensitivity, In Scouting, trigger thresholds are lowered along with event data size to keep data rates manageable. We have performed a search for pair-produced RPV gluinos decaying to three jets using the PF-Scouting dataset. We developed kinematic variables that were highly effective in differentiating the signal from QCD backgrounds and set extremely competitive limits on RPV gluinos. We are further improving this search by using machine learning(ML) techniques and developing new procedures to hunt for low mass boosted resonances using the Scouting datasets.
We are expanding our efforts to search for new resonances with the help of exciting and unique approaches using weakly supervised ML techniques. I will also talk about our efforts to improve searches for jet resonances by developing low-HT and displaced jet triggers at Level 1 for the HL-LHC.
