Online Program Home
My Program

Abstract Details

Activity Number: 256 - Contributed Poster Presentations: Section on Statistical Learning and Data Science
Type: Contributed
Date/Time: Monday, July 29, 2019 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistical Learning and Data Science
Abstract #307130
Title: Open Category Detection with PAC Guarantees
Author(s): Si Liu* and Risheek Garrepalli and Thomas G. Dietterich and Alan Fern and Dan Hendrycks
Companies: Oregon State University and Oregon State University and Oregon State University and Oregon State University and UC Berkeley
Keywords: Anomaly Detection; Open Category Detection

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 work, 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 of alien examples. Under the assumption that we know an upper bound on the proportion, we develop an algorithm with PAC-style guarantees on the alien detection rate, while aiming to minimize false alarms.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program