Timezone: »

 
Oral
POEM: Out-of-Distribution Detection with Posterior Sampling
Yifei Ming · Ying Fan · Sharon Li

Tue Jul 19 01:15 PM -- 01:35 PM (PDT) @ Hall F
in SA: Trustworthy Machine Learning »

Out-of-distribution (OOD) detection is indispensable for machine learning models deployed in the open world. Recently, the use of an auxiliary outlier dataset during training (also known as outlier exposure) has shown promising performance. As the sample space for potential OOD data can be prohibitively large, sampling informative outliers is essential. In this work, we propose a novel posterior sampling based outlier mining framework, POEM, which facilitates efficient use of outlier data and promotes learning a compact decision boundary between ID and OOD data for improved detection. We show that POEM establishes state-of-the-art performance on common benchmarks. Compared to the current best method that uses a greedy sampling strategy, POEM improves the relative performance by 42.0% and 24.2% (FPR95) on CIFAR-10 and CIFAR-100, respectively. We further provide theoretical insights on the effectiveness of POEM for OOD detection.

Keywords: SA: Trustworthy Machine Learning

Author Information

Yifei Ming (University of Wisconsin-Madison)
Ying Fan (University of Wisconsin-Madison)
Sharon Li (University of Wisconsin-Madison)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors