BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:AI-Driven Physics Instrument Design: From Reinforcement Learning t
 o Large Language Models
DTSTART:20260422T090000Z
DTEND:20260422T100000Z
DTSTAMP:20260423T180600Z
UID:indico-event-1648013@indico.cern.ch
CONTACT:EP-seminars.colloquia@cern.ch
DESCRIPTION:Speakers: Shah Rukh Qasim (University of Zurich (CH))\n\nDesig
 ning modern physics detectors is a high-dimensional\, combinatorial proble
 m that has traditionally relied on expert intuition and hand-tuned optimiz
 ation. This seminar presents two complementary machine-learning approaches
  that automate this process: reinforcement learning (RL) and large languag
 e models (LLMs).Reinforcement Learning  agents can explore complex design
  spaces without fixed parameterizations\, producing competitive detector l
 ayouts for tasks such as calorimeter segmentation and tracker placement. P
 retrained Large Language Models can  generate valid detector configuratio
 ns under the same simulation and reward framework. While RL achieves stron
 ger optimization performance\, LLMs reliably produce feasible\, resource-a
 ware designs and serve as high-level planners that can guide or structure 
 the search.Together\, these works point toward hybrid\, closed-loop workfl
 ows that combine conceptual planning by LLMs with fine-grained optimizatio
 n by RL to accelerate the design of next-generation physics instruments.\n
 Bio: Shah Rukh Qasim is a researcher jointly affiliated with the Physik-In
 stitut and the Department of Mathematical Modeling and Machine Learning at
  the University of Zurich. With a background in computer science\, he prev
 iously worked on particle reconstruction in high-energy physics during his
  Ph.D. at CERN with dynamic graph neural networks. His current work focuse
 s on applying machine learning—particularly reinforcement learning—to 
 detector and system design in particle physics\, including a core role in 
 the design optimization of the muon shield for the SHiP experiment\, along
 side broader interdisciplinary ML-driven research projects in epidemiology
  and supply chains.\n \ncoffee will be served at 10h30\n \n\nhttps://ind
 ico.cern.ch/event/1648013/\n\nZoom: https://cern.zoom.us/j/98545267593?pwd
 =akZWdmlyK01zbjFQa0x2c2ZXWW9ydz09
LOCATION:40/S2-A01 - Salle Anderson (CERN)
URL:https://indico.cern.ch/event/1648013/
END:VEVENT
END:VCALENDAR