BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260416T055736EDT-6784nsIoex@132.216.98.100
DTSTAMP:20260416T095736Z
DESCRIPTION:Title: Uncertainty quantification for black-box models with con
 ditional guarantees\n\nAbstract:\n\nA central problem in the uncertainty q
 uantification literature is designing methods that are both distribution-f
 ree and individualized to the test sample at hand. Prior work has shown th
 at it is impossible to achieve finite-sample conditional validity without 
 modelling assumptions. Thus\, canonical methods in\, e.g.\, the conformal 
 inference literature\, typically only issue marginal guarantees over a ran
 dom draw of the test covariates. In this talk\, I will outline a framework
  that bridges this gap by recasting the conditional objective as a set of 
 robustness criteria over a class of covariate shifts. By relaxing the targ
 et class of covariate shifts\, I will define a spectrum of problems that r
 ange between marginal and exact conditional validity and give methods that
  provide precise guarantees in between these extremes. This framework has 
 broad applications and I will show how it can be used to construct predict
 ion sets around the outputs of black-box regression models and filter out 
 false information from the responses of large language models. This talk i
 s based on joint work with John Cherian and Emmanuel Candès.\n\n🔗 Zoom: ht
 tps://mcgill.zoom.us/j/85280629047\n
DTSTART:20251203T163000Z
DTEND:20251203T173000Z
LOCATION:Room 1104\, Burnside Hall\, CA\, QC\, Montreal\, H3A 0B9\, 805 rue
  Sherbrooke Ouest
SUMMARY:Isaac Gibbs (University of California\, Berkley) 
URL:https://www.mcgill.ca/mathstat/channels/event/isaac-gibbs-university-ca
 lifornia-berkley-369347
END:VEVENT
END:VCALENDAR