Calibrating prediction uncertainty with validity guarantees

Add to your list(s) Download to your calendar using vCal

• Vladimir Vovk (Royal Holloway, University of London)
• Monday 02 June 2025, 14:00-15:00
• Seminar Room 1, Newton Institute.

If you have a question about this talk, please contact nobody.

RCLW02 - Calibrating prediction uncertainty : statistics and machine learning perspectives

My plan is to review several related methods for calibrating prediction uncertainty that enjoy various properties of validity under assumptions that are standard in mainstream machine learning. The method with the simplest validity guarantees is conformal prediction, which produces prediction sets with a given bound on the probability of error. Venn prediction and conformal predictive distributions are methods of probabilistic prediction that work in the case of classification and regression, respectively; their validity guarantees are more complicated but still natural.

This talk is part of the Isaac Newton Institute Seminar Series series.

This talk is included in these lists:

• All CMS events
• Featured lists
• INI info aggregator
• Isaac Newton Institute Seminar Series
• School of Physical Sciences
• Seminar Room 1, Newton Institute
• bld31

Note that ex-directory lists are not shown.