User Tools

Site Tools


Statistics Seminar
Department of Mathematical Sciences

DATE:Thursday, November 14, 2019
TIME:1:15pm – 2:15pm
SPEAKER:Baozhen Wang, Binghamton University
TITLE:Conformal prediction and the limit of distribution-free conditional predictive inference


Conformal prediction uses past experience to determine precise levels of confidence in new predictions. The proposed methodology allows for the construction of a distribution-free prediction bands for the response variable using any algorithm. The resulting prediction bands preserves marginal coverage guarantee, where predictive coverage holds on average over all possible test points, but is not sufficient for many practical applications. We aim to produce predictive conditional coverage rather than marginally with certain types of constrains and assumptions, while still being possible to achieve in a distribution-free setting.

seminars/stat/191114.txt · Last modified: 2019/11/13 12:58 by qyu