Open Category Detection with PAC Guarantees

TL;DR

This paper introduces a PAC-guaranteed algorithm for open category detection that effectively identifies alien instances in contaminated datasets, addressing key theoretical and empirical gaps in the field.

Contribution

It proposes the first algorithm with PAC guarantees for alien detection in contaminated training data, balancing detection accuracy and false alarms.

Findings

Algorithm achieves PAC-style guarantees on detection rate

Empirical results validate effectiveness on synthetic and benchmark datasets

Provides a baseline for future research in open category detection

Abstract

Open category detection is the problem of detecting "alien" test instances that belong to categories or classes that were not present in the training data. In many applications, reliably detecting such aliens is central to ensuring the safety and accuracy of test set predictions. Unfortunately, there are no algorithms that provide theoretical guarantees on their ability to detect aliens under general assumptions. Further, while there are algorithms for open category detection, there are few empirical results that directly report alien detection rates. Thus, there are significant theoretical and empirical gaps in our understanding of open category detection. In this paper, we take a step toward addressing this gap by studying a simple, but practically-relevant variant of open category detection. In our setting, we are provided with a "clean" training set that contains only the target categories of interest and an unlabeled "contaminated" training set that contains a fraction $\alpha$ of alien examples. Under the assumption that we know an upper bound on $\alpha$, we develop an algorithm with PAC-style guarantees on the alien detection rate, while aiming to minimize false alarms. Empirical results on synthetic and standard benchmark datasets demonstrate the regimes in which the algorithm can be effective and provide a baseline for further advancements.