Speaker: , KTH
Title: Conformal Prediction
Conformal prediction (CP) is a framework for quantifying the uncertainty of predictions. The framework, which can be used with any standard learning algorithm, allows the probability of making incorrect predictions to be bounded by a user-provided confidence threshold. In this talk, I will briefly introduce the framework and illustrate its use in conjunction with both interpretable models, such as decision trees, and highly predictive models, such as random forests.