Jianing Zhu, Geng Yu, Jiangchao Yao, Tongliang Liu, Gang Niu...
TL;DR通过生成多样化的离群点进行外部样本检测,提高模型的可靠性和效果。
Abstract
Out-of-distribution (OOD) detection is important for deploying reliable
machine learning models on real-world applications. Recent advances in outlier
exposure have shown promising results on ood detection via fine-tuning model
with informatively sampled auxiliary outliers. However, pr