BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Date iCal//NONSGML kigkonsult.se iCalcreator 2.20.2//
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:STANDARD
DTSTART:20251102T020000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20250309T020000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
RDATE:20260308T020000
TZNAME:EDT
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:calendar.3857.events_uoft_date.0@www.statistics.utoronto.ca
CREATED:20250925T173625Z
DESCRIPTION:\nWhen and Where: \nThursday, October 09, 2025 11:00 am to 1:
 00 pm \n 9014 \n 9th Floor 700 University Ave, Toronto, Ontario, M5G 1Z
 5 \n\nSpeakers \nXiao Wang, Purdue University \n\nDescription: \n Neural 
 Amortized Bayesian ComputationGiven an observation, how can we perform Ba
 yesian inference when the likelihood is intractable? Approximate Bayesian 
 Computation (ABC) provides a simple and intuitive solution with exactness 
 guarantees. However, its performance deteriorates rapidly as the paramete
 r dimension increases, a limitation known as the curse of dimensionality.
  We propose Neural Amortized Bayesian Computation (NABC), an extension of
  the ABC framework that alleviates this issue significantly while preservi
 ng its structure and exactness. NABC employs deep neural networks to captu
 re and remove the dependence between data and parameters for efficient inf
 erence. With such exactness, this approach not only outperforms state-of-
 the-art simulation-based inference methods, such as Neural Posterior Esti
 mation and Neural Likelihood Estimation, in amortized settings, but also
  achieves these improvements at substantially lower computational cost. We
  demonstrate its effectiveness on a range of benchmark studies reflecting 
 practical scenarios, and further extend its use to exact Bayesian inferen
 ce for differentially privatized data.BIO: Dr. Xiao Wang is Head and J.O. 
 Berger and M.E. Bock Professor of Statistics at Purdue University. He earn
 ed his Ph.D. from the University of Michigan, and his research centers on
  machine learning, nonparametric statistics, and functional data analysi
 s with particular emphasis on developing methods for high-dimensional and 
 complex data. His work has been featured in leading statistical journals a
 nd machine learning conferences, and he is a fellow of the Institute of M
 athematical Statistics (IMS) and the American Statistical Association (ASA
 ). He currently serves as an associate editor for JASA, Technometrics, a
 nd Lifetime Data Analysis. \n9th Floor 700 University Ave, Toronto, Onta
 rio, M5G 1Z5 \n\nCategories \n Statistics Colloquium \n\nAudiences \n Alu
 mni and FriendsFacultyGraduate StudentsPostdoctoral Fellows
DTSTART;TZID=America/New_York:20251009T110000
DTEND;TZID=America/New_York:20251009T130000
LAST-MODIFIED:20250925T175437Z
LOCATION:9th Floor 700 University Ave, Toronto, Ontario, M5G 1Z5
SUMMARY:Xiao Wang: Neural Amortized Bayesian Computation
URL;TYPE=URI:https://www.statistics.utoronto.ca/events/xiao-wang-neural-amo
 rtized-bayesian-computation-0
END:VEVENT
END:VCALENDAR