BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//jEvents 2.0 for Joomla//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:STANDARD
DTSTART:20241103T010000
RDATE:20250309T030000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:America/New_York EST
END:STANDARD
BEGIN:STANDARD
DTSTART:20251102T010000
RDATE:20260308T030000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:America/New_York EST
END:STANDARD
BEGIN:STANDARD
DTSTART:20261101T010000
RDATE:20270314T030000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:America/New_York EST
END:STANDARD
BEGIN:STANDARD
DTSTART:20271107T010000
RDATE:20280312T030000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:America/New_York EST
END:STANDARD
BEGIN:STANDARD
DTSTART:20281105T010000
RDATE:20290311T030000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:America/New_York EST
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20240916T103000
RDATE:20241103T010000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:America/New_York EDT
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:20250309T030000
RDATE:20251102T010000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:America/New_York EDT
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:20260308T030000
RDATE:20261101T010000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:America/New_York EDT
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:20270314T030000
RDATE:20271107T010000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:America/New_York EDT
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:20280312T030000
RDATE:20281105T010000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:America/New_York EDT
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:8d18df038a59510a2e903f3707e580f0
CATEGORIES:Mathematical Physics Seminar
CREATED:20250901T002555
SUMMARY:Webinar: Giulio Biroli - Generative AI and Diffusion Models: a statistical physics approach
LOCATION:Zoom
DESCRIPTION:<p style="text-align: center;"><strong>Giulio Biroli – Ecole Normale Supéri
 eure</strong></p><p style="text-align: center;"><strong>&nbsp;</strong></p>
 <p style="text-align: center;"><strong>&nbsp;</strong></p><p style="text-al
 ign: center;"><strong>Wednesday,&nbsp;September 17th,&nbsp;2025</strong></p
 ><p style="text-align: center;"><strong>Zoom opens: 10:30AM DST</strong></p
 ><p style="text-align: center;"><strong>Seminar begins: 10:45AM DST</strong
 ></p><p style="text-align: center;"><strong>&nbsp;</strong></p><p style="te
 xt-align: center;"><strong>Generative AI and Diffusion Models: a statistica
 l physics approach</strong></p><p>Generative AI represents a groundbreaking
  development within the broader “Machine Learning Revolution,” significantl
 y influencing technology, science, and society. In this talk, I will focus 
 on the state-of-the-art “diffusion models,” which are currently used to gen
 erate images, videos, and sounds. They are fascinating algorithms for physi
 cists, as they are very much connected to concepts from stochastic thermody
 namics, particularly time-reversed Langevin dynamics. Diffusion models init
 iate from a simple white noise input and evolve it through a Langevin proce
 ss to generate complex outputs such as images, videos, and sounds. I will s
 how that statistical physics provides guiding principles and methods to cha
 racterise this generation process. Specifically, I will discuss how phenome
 na such as the transition from memorization to generalization and the emerg
 ence of data-structure can be understood through the lens of symmetry break
 ing, phase transitions, and disordered systems.&nbsp;</p>
CONTACT:Giulio Biroli
DTSTAMP:20260827T020848
DTSTART;TZID=America/New_York:20250917T103000
DTEND;TZID=America/New_York:20250917T120000
SEQUENCE:0
TRANSP:OPAQUE
END:VEVENT
END:VCALENDAR