BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:NGT-WHISE Session 3
DTSTART:20260703T070000Z
DTEND:20260703T090000Z
DTSTAMP:20260820T081600Z
UID:indico-event-1704853@indico.cern.ch
CONTACT:ioannis.xiotidis@cern.ch\;maciej.glowacki@cern.ch\;alex.lasa.lamar
 ca@cern.ch
DESCRIPTION:Description\nThis NGT Whiteboard Session will focus on applyin
 g tools from theoretical physics\, geometry and topology towards understan
 ding the propagation of information through Neural Networks as well as und
 erstanding their key properties such as generalisability and robustness.\n
 This session will be organised on the occasion of Prof. Tolga Birdal’s v
 isit to CERN for the CERN STEAM Academy Seminar on Topological Deep Learni
 ng for the Next Generation of AI4Science.\nFollowing his seminar\, this de
 dicated NGT-WHISE session will provide an open space for further discussio
 n and making connections to research directions at CERN.\nTopological Deep
  Learning extends machine learning beyond graphs\, enabling models to repr
 esent higher-order interactions\, motifs\, cycles\, surfaces\, and richer 
 topological structures.\nAs with all NGT WHISE sessions\, the format will 
 be intentionally lightweight\, with minimal slides and a strong focus on o
 pen discussion and whiteboard-style exchange.\nCoffee and snacks will be p
 rovided.\nGuiding question\nHow can we guide Neural Networks towards learn
 ing maximally informative representation for optimal data acquisition and 
 insight?\nFormat\nThis is an informal discussion session aimed at explorin
 g early-stage ideas\, possible research directions\, and cross-WP connecti
 ons. Participants are encouraged to join actively\, ask questions\, and co
 ntribute perspectives from their own work.\nGuest\nProf. Tolga BirdalImper
 ial College London\nRegistration\nPlease register so that we can estimate 
 the number of participants.\n\nhttps://indico.cern.ch/event/1704853/
LOCATION:40/5-A01 (CERN)
URL:https://indico.cern.ch/event/1704853/
END:VEVENT
END:VCALENDAR