Deep Generative Classifier for Out-of-distribution Sample Detection

Dongha Lee, Sehun Yu, Hwanjo Yu

Keywords: Out-of-distribution Detection, Generative Classifier, Deep Neural Networks, Multi-class Classification, Gaussian Discriminant Analysis

TL;DR: This paper proposes a deep generative classifier which is effective to detect out-of-distribution samples as well as classify in-distribution samples, by integrating the concept of Gaussian discriminant analysis into deep neural networks.

Abstract: The capability of reliably detecting out-of-distribution samples is one of the key factors in deploying a good classifier, as the test distribution always does not match with the training distribution in most real-world applications. In this work, we propose a deep generative classifier which is effective to detect out-of-distribution samples as well as classify in-distribution samples, by integrating the concept of Gaussian discriminant analysis into deep neural networks. Unlike the discriminative (or softmax) classifier that only focuses on the decision boundary partitioning its latent space into multiple regions, our generative classifier aims to explicitly model class-conditional distributions as separable Gaussian distributions. Thereby, we can define the confidence score by the distance between a test sample and the center of each distribution. Our empirical evaluation on multi-class images and tabular data demonstrate that the generative classifier achieves the best performances in distinguishing out-of-distribution samples, and also it can be generalized well for various types of deep neural networks.