{"type":"ScholarlyArticle","title":"Robustness of Autoencoders for Anomaly Detection Under Adversarial Impact","authors":[{"name":"Adam Goodge","slug":"adam-goodge","given":"Adam","family":"Goodge","affiliation":"School of Computing , National University of Singapore; Institute for Infocomm Research , A*STAR , Singapore"},{"name":"Bryan Hooi","slug":"bryan-hooi","given":"Bryan","family":"Hooi","affiliation":"School of Computing , National University of Singapore; Institute of Data Science , National University of Singapore"},{"name":"See Kiong Ng","slug":"see-kiong-ng","given":"See Kiong","family":"Ng","affiliation":"School of Computing , National University of Singapore; Institute of Data Science , National University of Singapore"},{"name":"Wee Siong Ng","slug":"wee-siong-ng","given":"Wee Siong","family":"Ng","affiliation":"Institute for Infocomm Research , A*STAR , Singapore"}],"abstract":"Detecting anomalies is an important task in a wide variety of applications and domains. Deep learning methods have achieved state-of-the-art performance in anomaly detection in recent years; unsupervised methods being particularly popular. However, deep learning methods can be fragile to small perturbations in the input data. This can be exploited by an adversary to deliberately hinder model performance; an adversarial attack. This phenomena has been widely studied in the context of supervised image classification since its discovery, however such studies for an anomaly detection setting are sorely lacking. Moreover, the plethora of defense mechanisms that have been proposed are often not applicable to unsupervised anomaly detection models. In this work, we study the effect of adversarial attacks on the performance of anomaly-detecting autoencoders using real data from a Cyber physical system (CPS) testbed with intervals of controlled, physical attacks as anomalies. An adversary would attempt to disguise these points as normal through adversarial perturbations. To combat this, we propose the Approximate Projection Autoencoder (APAE), which incorporates two defenses against such attacks into a general autoencoder. One of these involves a novel technique to improve robustness under adversarial impact by optimising latent representations for better reconstruction outputs.","doi":"10.24963/ijcai.2020/173","url":"https://ijcai.org/proceedings/2020/173","pdf":"https://ijcai.org/proceedings/2020/0173.pdf","markdown":"https://ijcai.org/proceedings/2020/173.md","bibtex":"https://ijcai.org/proceedings/2020/bibtex/173","volume":{"slug":"2020","title":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2020","url":"https://ijcai.org/proceedings/2020/","year":2020,"location":"Yokohama","dates":"January 2021 (postponed from 11-17 July 2020)","editors":"Christian Bessiere","isbn":"978-0-9992411-6-5","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2020"},"section":"Data Mining","track":"Main track","number":173,"pages":{"first":1244,"last":1250},"keywords":["Machine Learning Applications: Applications of Unsupervised Learning","Data Mining: Mining Spatial, Temporal Data"],"videos":[{"label":"Video","url":"https://videos.ijcai20.org/24678/file_24678.m3u8","length_sec":273,"provider":"hls"}],"references":[{"position":1,"raw":"[An, 2015] Jinwon An. Variational Autoencoder based Anomaly Detection using Reconstruction Probability. In SNU Data Mining Center 2015-2 Special Lecture on IE, 2015.","title":"Jinwon An. Variational Autoencoder based Anomaly Detection using Reconstruction Probability","authors":"An","year":"2015","venue":"SNU Data Mining Center 2015-2 Special Lecture on IE","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Bo et al., 2018] Zong Bo, Qi Song, and H Chen. Deep autoencoding gaussian mixture model for unsuper- visedanomaly detection. 2018.","title":"Deep autoencoding gaussian mixture model for unsupervisedanomaly detection","authors":"Zong Bo, Qi Song, H Chen","year":"2018","venue":null,"doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Bolton and Hand, 1999] Richard Bolton and David Hand. Unsupervised profiling methods for fraud detection. In Proceedings of the Conference on Credit Scoring and Credit Control VII, 1999.","title":"Unsupervised profiling methods for fraud detection","authors":"Richard Bolton, David Hand","year":"1999","venue":"Proceedings of the Conference on Credit Scoring and Credit Control VII","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Chauhan and Vig, 2015] Sucheta Chauhan and Lovekesh Vig. Anomaly detection in ECG time signals via deep long short-term memory networks. In 2015 IEEE Inter- national Conference on Data Science and Advanced Ana- lytics (DSAA), pages 1-7, 2015.","title":"Anomaly detection in ECG time signals via deep long short-term memory networks","authors":"Vig Chauhan, Sucheta Chauhan, Lovekesh Vig","year":"2015","venue":"IEEE International Conference on Data Science and Advanced Analytics (DSAA)","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Chen et al., 2001] Yunqiang Chen, Xiang Sean Zhou, and Thomas S Huang. One-class SVM for learning in image retrieval. In International Conference on Image Process- ing, pages 34-37. Citeseer, 2001.","title":"One-class SVM for learning in image retrieval","authors":"Chen","year":"2001","venue":"International Conference on Image Processing","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Chen et al., 2018] Zhaomin Chen, Chai Kiat Yeo, Bu Sung Lee, and Chiew Tong Lau. Autoencoder-based network anomaly detection. In 2018 Wireless Telecommunications Symposium (WTS), pages 1-5, 2018.","title":"Autoencoder-based network anomaly detection","authors":"Zhaomin Chen, Kiat Chai, Bu Yeo, Sung Lee, Chiew Tong Lau","year":"2018","venue":"Wireless Telecommunications Symposium (WTS)","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Dau et al., 2014] Hoang Anh Dau, Vic Ciesielski, and Andy Song. Anomaly detection using replicator neural networks trained on examples of one class. In Asia-Pacific Confer- ence on Simulated Evolution and Learning, pages 311- 322. Springer, 2014.","title":"Anomaly detection using replicator neural networks trained on examples of one class","authors":"Anh Hoang, Vic Dau, Andy Ciesielski, Song","year":"2014","venue":"Asia-Pacific Conference on Simulated Evolution and Learning","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Eskin et al., 2002] Eleazar Eskin, Andrew Arnold, Michael Prerau, Leonid Portnoy, and Sal Stolfo. A geometric framework for unsupervised anomaly detection. In Ap- plications of data mining in computer security, pages 77- 101. Springer, 2002.","title":"A geometric framework for unsupervised anomaly detection","authors":"Eskin","year":"2002","venue":"Applications of data mining in computer security","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Feng and Han, 2015] Wenhui Feng and Chongzhao Han. A Novel Approach for Trajectory Feature Representation and Anomalous Trajectory Detection. International Con- ference on Information Fusion, pages 1093-1099, 2015.","title":"A Novel Approach for Trajectory Feature Representation and Anomalous Trajectory Detection","authors":"Han ; Feng, Wenhui Feng, Chongzhao Han","year":"2015","venue":"International Conference on Information Fusion","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Goldstein and Uchida, 2016] Markus Goldstein and Seiichi Uchida. A Comparative Evaluation of Unsupervised Anomaly Detection Algorithms for Multivariate Data. PLOS ONE, 11(4):e0152173, apr 2016.","title":"A Comparative Evaluation of Unsupervised Anomaly Detection Algorithms for Multivariate Data","authors":"Uchida Goldstein, Markus Goldstein, Seiichi Uchida","year":"2016","venue":"PLOS ONE","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Goodfellow et al., 2014] Ian J Goodfellow, Jonathon Shlens, and Christian Szegedy. Explaining and harnessing adversarial examples. arXiv preprint arXiv:1412.6572, 2014.","title":"Explaining and harnessing adversarial examples","authors":"Ian J Goodfellow, Jonathon Shlens, Christian Szegedy","year":"2014","venue":null,"doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Hawkins et al., 2002] Simon Hawkins, Hongxing He, Gra- ham Williams, and Rohan Baxter. Outlier detection using replicator neural networks. DaWaK2002. LNCS, 2454:170-180, 2002.","title":"Outlier detection using replicator neural networks","authors":"Simon Hawkins, Hongxing He, Graham Williams, Rohan Baxter","year":"2002","venue":"DaWaK2002","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Hawkins, 1980] Douglas Hawkins. Identification of Out- liers. Chapman and Hall, London, 1980.","title":"Identification of Outliers","authors":"Douglas Hawkins","year":"1980","venue":null,"doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Hu et al., 2003] Wenjie Hu, Yihua Liao, and V Rao Vemuri. Robust anomaly detection using support vector machines. In Proceedings of the international conference on machine learning, pages 282-289, 2003. iTrust Centre for Research in Cyber Security, 2019] iTrust Centre for Research in Cyber Security. Datasets, 2019.","title":"Robust anomaly detection using support vector machines","authors":"Wenjie Hu, Yihua Liao, V Rao Vemuri","year":"2003","venue":"Proceedings of the international conference on machine learning","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Kloft and Laskov, 2010] Marius Kloft and Pavel Laskov. Online Anomaly Detection under Adversarial Impact. In Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, pages 405-412, 2010.","title":"Marius Kloft and Pavel Laskov. Online Anomaly Detection under Adversarial Impact","authors":"Laskov Kloft","year":"2010","venue":"Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Kurakin et al., 2017] Alexey Kurakin, Ian Goodfellow, and Samy Bengio. Adversarial machine learning at scale. In ICLR 2017, 2017.","title":"Adversarial machine learning at scale","authors":"Alexey Kurakin, Ian Goodfellow, Samy Bengio","year":"2017","venue":"ICLR 2017","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Liu et al., 2019] Lingbo Liu, Zhilin Qiu, Guanbin Li, Qing Wang, Wanli Ouyang, and Liang Lin. Contextualized Spatial-Temporal Network for Taxi Origin-Destination Demand Prediction. IEEE Transactions on Intelligent Transportation Systems, PP:1-13, 2019.","title":"Contextualized Spatial-Temporal Network for Taxi Origin-Destination Demand Prediction","authors":"Lingbo Liu, Zhilin Qiu, Guanbin Li, Qing Wang, Wanli Ouyang, Liang Lin","year":"2019","venue":"IEEE Transactions on Intelligent Transportation Systems","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Malhotra et al., 2015] Pankaj Malhotra, Lovekesh Vig, Gautam Shroff, and Puneet Agarwal. Long short term memory networks for anomaly detection in time series. In ESANN 2015 Proceedings, European Symposium on Ar- tificial Neural Networks, Computational Intelligence and Machine Learning, pages 89-94. Presses universitaires de Louvain, 2015.","title":"Long short term memory networks for anomaly detection in time series","authors":"Pankaj Malhotra, Lovekesh Vig, Gautam Shroff, Puneet Agarwal","year":"2015","venue":"ESANN 2015 Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Paudice et al., 2018] Andrea Paudice, Luis Munoz- Gonzalez, Andras Gyorgy, and Emil Lupu. Detection of Adversarial Training Examples in Poisoning Attacks through the Anomaly Detection. 2018.","title":"Detection of Adversarial Training Examples in Poisoning Attacks through the Anomaly Detection","authors":"Andrea Paudice, Luis Munoz-Gonzalez, Andras Gyorgy, Emil Lupu","year":"2018","venue":null,"doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Sakurada and Yairi, 2014] Mayu Sakurada and Takehisa Yairi. Anomaly detection using autoencoders with non- linear dimensionality reduction. In Proceedings of the MLSDA 2014 2nd Workshop on Machine Learning for Sensory Data Analysis, page 4. ACM, 2014.","title":"Anomaly detection using autoencoders with nonlinear dimensionality reduction","authors":"Yairi ; ] Mayu Sakurada, Takehisa Sakurada, Yairi","year":"2014","venue":"Proceedings of the MLSDA 2014 2nd Workshop on Machine Learning for Sensory Data Analysis","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Spence et al., 2001] Clay Spence, Lucas Parra, and Paul Sajda. Detection, synthesis and compression in mammo- graphic image analysis with a hierarchical image probabil- ity model. In Proceedings of the IEEE Workshop on Math- ematical Methods in Biomedical Image Analysis, 2001.","title":"Detection, synthesis and compression in mammographic image analysis with a hierarchical image probability model","authors":"Spence","year":"2001","venue":"Proceedings of the IEEE Workshop on Mathematical Methods in Biomedical Image Analysis","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Szegedy et al., 2013] Christian Szegedy, Wojciech Zaremba, Ilya Sutskever, Joan Bruna, Dumitru Erhan, Ian Goodfellow, and Rob Fergus. Intriguing properties of neural networks. arXiv preprint arXiv:1312.6199, 2013.","title":"Intriguing properties of neural networks","authors":"Szegedy","year":"2013","venue":null,"doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Zhang et al., 2001] Zheng Zhang, Jun Li, C.N. Manikopou- los, Jay Jorgenson, and Jose Ucles. HIDE: a hierarchical network intrusion detection system using statistical pre- processing and neural network classification. In Proc. IEEE Workshop on Information Assurance and Security, pages 85-90, 2001.","title":"HIDE: a hierarchical network intrusion detection system using statistical preprocessing and neural network classification","authors":"Zheng Zhang, Jun Li, C N Manikopoulos, Jay Jorgenson, Jose Ucles","year":"2001","venue":"Proc. IEEE Workshop on Information Assurance and Security","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Zhou and Paffenroth, 2017] Chong Zhou and Randy C Paf- fenroth. Anomaly detection with robust deep autoen- coders. In Proceedings of the 23rd ACM SIGKDD Inter- national Conference on Knowledge Discovery and Data Mining, pages 665-674. ACM, 2017.","title":"Anomaly detection with robust deep autoencoders","authors":"Chong Zhou, Randy C Paffenroth","year":"2017","venue":"Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}