BEGIN:VCALENDAR
PRODID:-//Columba Systems Ltd//NONSGML CPNG/SpringViewer/ICal Output/3.3-
 M3//EN
VERSION:2.0
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
DTSTAMP:20250304T165016Z
DTSTART:20250312T110000Z
SUMMARY:AI-Fun with ELLIS Seminar | Siu Lun Chau: Credal Two-sample Tests
 : How to Test Hypothesis under Dataset Uncertainty
UID:{http://www.columbasystems.com/customers/uom/gpp/eventid/}n1sm-m7umph
 7q-4h23l8
DESCRIPTION:March’s AI-Fun with ELLIS Seminar is with Siu Lun (Alan) Chau
  from CISPA Helmholtz Center for Information Security.\n\nIn-person atte
 ndance is recommended but for anyone who cannot be physically present\, 
 there is a zoom link included below.\n\nTitle: \nCredal Two-sample Tests
 : How to Test Hypothesis under Dataset Uncertainty\n \nAbstract:\nDatase
 t-level uncertainty arises when there is ambiguity about the true distri
 bution of interest. This can occur in scenarios involving distribution s
 hifts\, when datasets act as proxies\, or when multiple datasets are col
 lected under varying conditions. In this talk\, we introduce credal two-
 sample testing\, a novel hypothesis testing framework that compares cred
 al sets—convex sets of probability measures that capture both aleatoric 
 and epistemic uncertainty. Unlike classical two-sample tests\, which com
 pare precise distributions\, our approach supports reasoning about equal
 ity\, inclusion\, intersection\, and mutual exclusivity between credal s
 ets. This generalisation allows hypothesis testing to account for episte
 mic uncertainty directly\, addressing challenges posed by partial ignora
 nce stemming from dataset-level ambiguity. We formalise these tests as t
 wo-sample problems with nuisance parameters\, and we propose the first p
 ermutation-based solution for this class of problems. We also present pr
 actical\, kernel-based implementations\, enabling the application of cre
 dal two-sample testing to real-world data analysis tasks.\n \nProfile:\n
 Siu Lun (Alan) Chau is an incoming Assistant Professor in Statistical Ma
 chine Learning at Nanyang Technological University\, Singapore. His rese
 arch focuses on uncertainty-aware machine learning\, explainable machine
  learning\, and preference-integrated learning. Currently\, he is a Post
 doctoral Researcher at the Rational Intelligence Lab at CISPA Helmholtz 
 Center for Information Security in Germany\, where he works with Dr. Kri
 kamol Muandet on imprecise-probabilistic machine learning. Siu Lun compl
 eted his DPhil in Statistics under the supervision of Prof. Dino Sejdino
 vic at the University of Oxford\, where his research explored the integr
 ation of kernel methods and Gaussian processes\, with a focus on their a
 pplications to trustworthy machine learning.\n\nWe hope to see you there
 !\n\nBest wishes\,\n\nAI-Fundamentals\nELLIS Manchester\n
STATUS:TENTATIVE
TRANSP:TRANSPARENT
CLASS:PUBLIC
LOCATION:3A.057\, Nancy Rothwell Building\, Booth St East\, Manchester
END:VEVENT
END:VCALENDAR