TY - GEN
T1 - Gradient-Based Novelty Detection Boosted by Self-Supervised Binary Classification
AU - Sun, Jingbo
AU - Yang, Li
AU - Zhang, Jiaxin
AU - Liu, Frank
AU - Halappanavar, Mahantesh
AU - Fan, Deliang
AU - Cao, Yu
N1 - Funding Information:
This research was supported in part by the U.S. Department of Energy, through the Office of Advanced Scientific Computing Research’s “Data-Driven Decision Control for Complex Systems (DnC2S)” project. It was also partially supported by C-BRIC, one of six centers in JUMP, a Semiconductor Research Corporation (SRC) program sponsored by DARPA. Pacific Northwest National Laboratory is operated by Battelle Memorial Institute for the U.S. Department of Energy under Contract No. DE-AC05-76RL01830. Oak Ridge National Laboratory is operated by UT-Battelle LLC for the U.S. Department of Energy under contract number DE-AC05-00OR22725.
Publisher Copyright:
Copyright © 2022, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.
PY - 2022/6/30
Y1 - 2022/6/30
N2 - Novelty detection aims to automatically identify out-of-distribution (OOD) data, without any prior knowledge of them. It is a critical step in data monitoring, behavior analysis and other applications, helping enable continual learning in the field. Conventional methods of OOD detection perform multi-variate analysis on an ensemble of data or features, and usually resort to the supervision with OOD data to improve the accuracy. In reality, such supervision is impractical as one cannot anticipate the anomalous data. In this paper, we propose a novel, self-supervised approach that does not rely on any pre-defined OOD data: (1) The new method evaluates the Mahalanobis distance of the gradients between the in-distribution and OOD data. (2) It is assisted by a self-supervised binary classifier to guide the label selection to generate the gradients, and maximize the Mahalanobis distance. In the evaluation with multiple datasets, such as CIFAR-10, CIFAR-100, SVHN and TinyImageNet, the proposed approach consistently outperforms state-of-the-art supervised and unsupervised methods in the area under the receiver operating characteristic (AUROC) and area under the precision-recall curve (AUPR) metrics. We further demonstrate that this detector is able to accurately learn one OOD class in continual learning.
AB - Novelty detection aims to automatically identify out-of-distribution (OOD) data, without any prior knowledge of them. It is a critical step in data monitoring, behavior analysis and other applications, helping enable continual learning in the field. Conventional methods of OOD detection perform multi-variate analysis on an ensemble of data or features, and usually resort to the supervision with OOD data to improve the accuracy. In reality, such supervision is impractical as one cannot anticipate the anomalous data. In this paper, we propose a novel, self-supervised approach that does not rely on any pre-defined OOD data: (1) The new method evaluates the Mahalanobis distance of the gradients between the in-distribution and OOD data. (2) It is assisted by a self-supervised binary classifier to guide the label selection to generate the gradients, and maximize the Mahalanobis distance. In the evaluation with multiple datasets, such as CIFAR-10, CIFAR-100, SVHN and TinyImageNet, the proposed approach consistently outperforms state-of-the-art supervised and unsupervised methods in the area under the receiver operating characteristic (AUROC) and area under the precision-recall curve (AUPR) metrics. We further demonstrate that this detector is able to accurately learn one OOD class in continual learning.
UR - http://www.scopus.com/inward/record.url?scp=85147657271&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85147657271&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85147657271
T3 - Proceedings of the 36th AAAI Conference on Artificial Intelligence, AAAI 2022
SP - 8370
EP - 8377
BT - AAAI-22 Technical Tracks 8
PB - Association for the Advancement of Artificial Intelligence
T2 - 36th AAAI Conference on Artificial Intelligence, AAAI 2022
Y2 - 22 February 2022 through 1 March 2022
ER -