BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Beyond Reconstruction: Machine Learning with Low-Level Detector In
 formation for New Physics Searches in ATLAS
DTSTART:20260818T193000Z
DTEND:20260818T203000Z
DTSTAMP:20260824T060300Z
UID:indico-event-10681@indico.slac.stanford.edu
DESCRIPTION:Speakers: Gabriel Matos (Columbia U)\n\nThe limitations of the
  Standard Model continue to motivate searches for physics beyond the Stand
 ard Model (BSM). Many BSM scenarios remain experimentally elusive\, often 
 because their signatures are challenging to identify with traditional reco
 nstruction techniques at collider experiments. In these instances\, low-le
 vel detector information\, representing the inputs to traditional reconstr
 uction algorithms\, can provide a richer representation of collision event
 s than conventional high-level physics objects. This talk explores how mac
 hine learning (ML) can exploit this low-level information to improve sensi
 tivity to various BSM signatures with the ATLAS Experiment. Several exampl
 es of this approach will illustrate the gains achievable in common collide
 r tasks\, including event reconstruction\, classification\, and anomaly de
 tection\, through the use of inputs at the particle\, track\, and calorime
 ter-cell levels. These will be discussed in the context of searches for a 
 range of BSM signatures\, including jet substructure from boosted new heav
 y particle decays\, semi-visible jets arising from a hidden dark sector\, 
 and highly collimated photon pairs produced from axion-like particle decay
 s. Together\, these examples demonstrate how low-level\, detector-driven M
 L can provide new tools for probing BSM physics across a diverse range of 
 production mechanisms and experimental signatures.\n\n\n\n\n\n\n\nhttps://
 stanford.zoom.us/j/98973156241?pwd=cEU5RFdlVXoyc0JTeTlDMkozKzQ5UT09\n\n\n\
 n\n\n\n\n\nhttps://indico.slac.stanford.edu/event/10681/
LOCATION:48/2-224 - Madrone (SLAC)
URL:https://indico.slac.stanford.edu/event/10681/
END:VEVENT
END:VCALENDAR