Feature Space Singularity for Out-of-Distribution Detection

11/30/2020
by   Haiwen Huang, et al.
9

Out-of-Distribution (OoD) detection is important for building safe artificial intelligence systems. However, current OoD detection methods still cannot meet the performance requirements for practical deployment. In this paper, we propose a simple yet effective algorithm based on a novel observation: in a trained neural network, OoD samples with bounded norms well concentrate in the feature space. We call the center of OoD features the Feature Space Singularity (FSS), and denote the distance of a sample feature to FSS as FSSD. Then, OoD samples can be identified by taking a threshold on the FSSD. Our analysis of the phenomenon reveals why our algorithm works. We demonstrate that our algorithm achieves state-of-the-art performance on various OoD detection benchmarks. Besides, FSSD also enjoys robustness to slight corruption in test data and can be further enhanced by ensembling. These make FSSD a promising algorithm to be employed in real world. We release our code at <https://github.com/megvii-research/FSSD_OoD_Detection>.

READ FULL TEXT

page 2

page 3

page 6

page 7

research
02/02/2022

VOS: Learning What You Don't Know by Virtual Outlier Synthesis

Out-of-distribution (OOD) detection has received much attention lately d...
research
04/13/2017

A Neural Model for User Geolocation and Lexical Dialectology

We propose a simple yet effective text- based user geolocation model bas...
research
06/26/2023

Beyond AUROC co. for evaluating out-of-distribution detection performance

While there has been a growing research interest in developing out-of-di...
research
10/27/2022

Rethinking the Reverse-engineering of Trojan Triggers

Deep Neural Networks are vulnerable to Trojan (or backdoor) attacks. Rev...
research
03/21/2022

ViM: Out-Of-Distribution with Virtual-logit Matching

Most of the existing Out-Of-Distribution (OOD) detection algorithms depe...
research
10/18/2021

Natural Attribute-based Shift Detection

Despite the impressive performance of deep networks in vision, language,...
research
09/20/2022

Extremely Simple Activation Shaping for Out-of-Distribution Detection

The separation between training and deployment of machine learning model...

Please sign up or login with your details

Forgot password? Click here to reset