Abstract
Out-of-Distribution (OoD) detectors based on AutoEncoder (AE) rely on an underlying assumption that an AE network cannot reconstruct OoD data as good as in-distribution (ID) data when it is constructed based on ID data only. However, this assumption may be violated in practice, resulting in a degradation in detection performance. Therefore, alleviating the factors violating this assumption can potentially improve the robustness of OoD performance. Our empirical studies also show that image complexity can be another factor hindering detection performance for AE-based detectors. To cater for these issues, we propose two OoD detectors LAMAE and LAMAE+. Both can be trained without the availability of any OoD-related data. The key idea is to regularize the AE network architecture with a classifier and a label-assisted memory to confine the reconstruction of OoD data while retaining the reconstruction ability for ID data. We also adjust the reconstruction error by taking image complexity into consideration. Experimental studies show that the proposed OoD detectors can perform well on a wider range of OoD scenarios.
Original language | English |
---|---|
Title of host publication | Machine Learning and Knowledge Discovery in Databases. Research Track : European Conference, ECML PKDD 2021, Bilbao, Spain, September 13–17, 2021, Proceedings, Part III |
Editors | Nuria OLIVER, Fernando PÉREZ-CRUZ, Stefan KRAMER, Jesse READ, Jose A. LOZANO |
Publisher | Springer Science and Business Media Deutschland GmbH |
Pages | 795-810 |
Number of pages | 16 |
ISBN (Electronic) | 9783030865238 |
ISBN (Print) | 9783030865221 |
DOIs | |
Publication status | Published - 2021 |
Externally published | Yes |
Event | European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2021 - Bilbao, Spain Duration: 13 Sept 2021 → 17 Sept 2021 |
Publication series
Name | Lecture Notes in Computer Science |
---|---|
Publisher | Springer |
ISSN (Print) | 0302-9743 |
ISSN (Electronic) | 1611-3349 |
Name | Lecture Notes in Artificial Intelligence |
---|---|
Publisher | Springer |
Volume | 12977 |
ISSN (Print) | 2945-9133 |
ISSN (Electronic) | 2945-9141 |
Public Lecture
Public Lecture | European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2021 |
---|---|
Abbreviated title | ECML PKDD 2021 |
Country/Territory | Spain |
City | Bilbao |
Period | 13/09/21 → 17/09/21 |
Bibliographical note
Publisher Copyright:© 2021, Springer Nature Switzerland AG.
Funding
This work was supported by National Natural Science Foundation of China (Grant No. 62002148), the Guangdong Provincial Key Laboratory (Grant No. 2020B121201001), the Program for Guangdong Introducing Innovative and Entrepreneurial Teams (Grant No. 2017ZT07X386), Shenzhen Science and Technology Program (Grant No. KQTD2016112514355531), Shenzhen Fundamental Research Program (Grant No. JCYJ20190809121403553), Research Institute of Trustworthy Autonomous Systems, and Huawei.