Timezone: »

 
Poster
Conformalized Quantile Regression
Yaniv Romano · Evan Patterson · Emmanuel Candes

Wed Dec 11 10:45 AM -- 12:45 PM (PST) @ East Exhibition Hall B + C #42
in Algorithms -- Uncertainty Estimation »

Conformal prediction is a technique for constructing prediction intervals that attain valid coverage in finite samples, without making distributional assumptions. Despite this appeal, existing conformal methods can be unnecessarily conservative because they form intervals of constant or weakly varying length across the input space. In this paper we propose a new method that is fully adaptive to heteroscedasticity. It combines conformal prediction with classical quantile regression, inheriting the advantages of both. We establish a theoretical guarantee of valid coverage, supplemented by extensive experiments on popular regression datasets. We compare the efficiency of conformalized quantile regression to other conformal methods, showing that our method tends to produce shorter intervals.

Keywords: Algorithms · Uncertainty Estimation · Algorithms -> Regression; Deep Learning -> Predictive Models; Deep Learning -> Supervised Deep Networks; Theory · Frequentist
[ Paper]  [ Poster] 

Author Information

Yaniv Romano (Stanford University)
Evan Patterson (Stanford University)
Emmanuel Candes (Stanford University)

More from the Same Authors