BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Beyond Conformal Prediction: Distribution-Free Uncertainty Quantif
 ication for Complex Machine Learning Tasks - Anastasios Angelopoulos\, PhD
  student at UC Berkeley
DTSTART:20220513T103000Z
DTEND:20220513T113000Z
UID:TALK174200@talks.cam.ac.uk
CONTACT:Adrian Weller
DESCRIPTION:As we begin deploying machine learning models in consequential
  settings like medical diagnostics or self-driving vehicles\, we need ways
  of knowing when the model may make a consequential error (for example\, t
 hat the car doesn't hit a human). I'll be discussing how to generate rigor
 ous\, finite-sample confidence intervals for any prediction task\, any mod
 el\, and any dataset\, for free. This will be a chalk talk. I will primari
 ly discuss a flexible method called Learn then Test that works for a large
  class of prediction problems including those with high-dimensional\, stru
 ctured outputs (e.g. instance segmentation\, multiclass or hierarchical cl
 assification\, protein folding\, and so on).
LOCATION:Hybrid meeting\, CBL seminar room\, and Zoom https://eng-cam.zoom
 .us/j/84829493214?pwd=HI3JJJ8nrt6O8AELEzbXvT8ZoBj51t.1
END:VEVENT
END:VCALENDAR