Efficiency criteria for conformal prediction, such as \emph{observed fuzziness} (i.e., the sum of p-values associated with false labels), are commonly used to \emph{evaluate} the performance of given conformal predictors. Here, we investigate whether it is possible to exploit efficiency criteria to \emph{learn} classifiers, both conformal predictors and point classifiers, by using such criteria as training objective functions. The proposed idea is implemented for the problem of binary classification of hand-written digits. By choosing a 1-dimensional model class (with one real-valued free parameter), we can solve the optimization problems through an (approximate) exhaustive search over (a discrete version of) the parameter space. Our empirical results suggest that conformal predictors trained by minimizing their observed fuzziness perform better than conformal predictors trained in the traditional way by minimizing the \emph{prediction error} of the corresponding point classifier. They also have a reasonable performance in terms of their prediction error on the test set.
[1]
Isidro Cortes-Ciriano,et al.
Deep Confidence: A Computationally Efficient Framework for Calculating Reliable Errors for Deep Neural Networks
,
2018,
Journal of chemical information and modeling.
[2]
Alexander Gammerman,et al.
Criteria of efficiency for set-valued classification
,
2017,
Annals of Mathematics and Artificial Intelligence.
[3]
W. Gasarch,et al.
The Book Review Column 1 Coverage Untyped Systems Simple Types Recursive Types Higher-order Systems General Impression 3 Organization, and Contents of the Book
,
2022
.
[4]
Jing Lei,et al.
Fast Exact Conformalization of Lasso using Piecewise Linear Homotopy
,
2017,
1708.00427.
[5]
Ryan J. Tibshirani,et al.
Predictive inference with the jackknife+
,
2019,
The Annals of Statistics.