BEGIN:VCALENDAR VERSION:2.0 PRODID:researchseminars.org CALSCALE:GREGORIAN X-WR-CALNAME:researchseminars.org BEGIN:VEVENT SUMMARY:Aurelien Dersy (Harvard) DTSTART:20251014T183000Z DTEND:20251014T193000Z DTSTAMP:20260423T005746Z UID:nhetc/123 DESCRIPTION:Title: Learning the Language of Scattering Amplitudes\nby Aurelien Dersy (Har vard) as part of NHETC Seminar\n\n\nAbstract\nMachine learning (ML) has gr own dramatically in high-energy physics in recent years. Tasks such as cla ssification\, regression\, anomaly detection\, and density estimation now benefit from modern ML approaches that often outperform traditional baseli nes. Far less attention\, however\, has gone to the theoretical side\, eve n though (symbolic) data is plentiful. In this talk\, I will discuss how m odern models can help bridge this gap and “learn the language” of scat tering amplitudes. In particular\, I will describe recent work on simplify ing amplitude expressions\, reconstructing S-matrix phases\, and recoverin g exact analytic formulas.\n LOCATION:https://researchseminars.org/talk/nhetc/123/ END:VEVENT END:VCALENDAR