Explainable Deep One-Class Classification

Deep one-class classification variants for anomaly detection learn a mapping that concentrates nominal samples in feature space causing anomalies to be mapped away. Because this transformation is highly non-linear, finding interpretations poses a significant challenge. In this paper we present an explainable deep one-class classification method, Fully Convolutional Data Description (FCDD), where the mapped samples are themselves also an explanation heatmap. FCDD yields competitive detection performance and provides reasonable explanations on common anomaly detection benchmarks with CIFAR-10 and ImageNet. On MVTec-AD, a recent manufacturing dataset offering ground-truth anomaly maps, FCDD meets the state of the art in an unsupervised setting, and outperforms its competitors in a semi-supervised setting. Finally, using FCDD's explanations we demonstrate the vulnerability of deep one-class classification models to spurious image features such as image watermarks.

[1]  Thomas G. Dietterich,et al.  Deep Anomaly Detection with Outlier Exposure , 2018, ICLR.

[2]  F. Y. Edgeworth,et al.  XLI. On discordant observations , 1887 .

[3]  Frederick R. Forst,et al.  On robust estimation of the location parameter , 1980 .

[4]  VARUN CHANDOLA,et al.  Anomaly detection: A survey , 2009, CSUR.

[5]  Raquel Urtasun,et al.  Understanding the Effective Receptive Field in Deep Convolutional Neural Networks , 2016, NIPS.

[6]  Mathieu Lamard,et al.  Multiple-Instance Learning for Anomaly Detection in Digital Mammography , 2016, IEEE Transactions on Medical Imaging.

[7]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[8]  Kent A. Spackman,et al.  Signal Detection Theory: Valuable Tools for Evaluating Inductive Learning , 1989, ML.

[9]  Zhenyu Li,et al.  Superpixel Masking and Inpainting for Self-Supervised Anomaly Detection , 2020, BMVC.

[10]  Wojciech Samek,et al.  Toward Interpretable Machine Learning: Transparent Deep Neural Networks and Beyond , 2020, ArXiv.

[11]  Carsten Steger,et al.  MVTec AD — A Comprehensive Real-World Dataset for Unsupervised Anomaly Detection , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[12]  Mahmood Fathy,et al.  Deep-anomaly: Fully convolutional neural network for fast anomaly detection in crowded scenes , 2016, Comput. Vis. Image Underst..

[13]  Georg Langs,et al.  Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery , 2017, IPMI.

[14]  Silvio Savarese,et al.  Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[15]  Geoffrey E. Hinton,et al.  On the importance of initialization and momentum in deep learning , 2013, ICML.

[16]  M. M. Moya,et al.  One-class classifier networks for target recognition applications , 1993 .

[17]  Bernhard Schölkopf,et al.  Estimating the Support of a High-Dimensional Distribution , 2001, Neural Computation.

[18]  Anna Shcherbina,et al.  Not Just a Black Box: Learning Important Features Through Propagating Activation Differences , 2016, ArXiv.

[19]  Chen Shen,et al.  Spatio-Temporal AutoEncoder for Video Anomaly Detection , 2017, ACM Multimedia.

[20]  Paolo Napoletano,et al.  Anomaly Detection in Nanofibrous Materials by CNN-Based Self-Similarity , 2018, Sensors.

[21]  Léon Bottou,et al.  Large-Scale Machine Learning with Stochastic Gradient Descent , 2010, COMPSTAT.

[22]  Attention Guided Anomaly Detection and Localization in Images , 2019, ArXiv.

[23]  Jun Cheng,et al.  Encoding Structure-Texture Relation with P-Net for Anomaly Detection in Retinal Images , 2020, ECCV.

[24]  Takehisa Yairi,et al.  Anomaly Detection Using Autoencoders with Nonlinear Dimensionality Reduction , 2014, MLSDA'14.

[25]  Alexander Binder,et al.  Analyzing Classifiers: Fisher Vectors and Deep Neural Networks , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[26]  Li Fei-Fei,et al.  ImageNet: A large-scale hierarchical image database , 2009, CVPR.

[27]  Dawn Song,et al.  Using Self-Supervised Learning Can Improve Model Robustness and Uncertainty , 2019, NeurIPS.

[28]  Andrew Zisserman,et al.  Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps , 2013, ICLR.

[29]  Mykel J. Kochenderfer,et al.  Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networks , 2017, CAV.

[30]  Cewu Lu,et al.  Inverse-Transform AutoEncoder for Anomaly Detection , 2019, ArXiv.

[31]  Ali Farhadi,et al.  You Only Look Once: Unified, Real-Time Object Detection , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[32]  Mohammad Hossein Jarrahi,et al.  Artificial intelligence and the future of work: Human-AI symbiosis in organizational decision making , 2018, Business Horizons.

[33]  Trevor Darrell,et al.  Fully Convolutional Networks for Semantic Segmentation , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[34]  Seunghoon Hong,et al.  Learning Deconvolution Network for Semantic Segmentation , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[35]  Marius Kloft,et al.  Self-Attentive, Multi-Context One-Class Classification for Unsupervised Anomaly Detection on Text , 2019, ACL.

[36]  Yedid Hoshen,et al.  Classification-Based Anomaly Detection for General Data , 2020, ICLR.

[37]  Alexander Binder,et al.  Deep Semi-Supervised Anomaly Detection , 2019, ICLR.

[38]  Antonio Torralba,et al.  Ieee Transactions on Pattern Analysis and Machine Intelligence 1 80 Million Tiny Images: a Large Dataset for Non-parametric Object and Scene Recognition , 2022 .

[39]  Gregory Cohen,et al.  EMNIST: Extending MNIST to handwritten letters , 2017, 2017 International Joint Conference on Neural Networks (IJCNN).

[40]  Harsha Vardhan Simhadri,et al.  DROCC: Deep Robust One-Class Classification , 2020, ICML.

[41]  Andreas Krause,et al.  Safe Model-based Reinforcement Learning with Stability Guarantees , 2017, NIPS.

[42]  Oriel Frigo,et al.  Iterative energy-based projection on a normal data manifold for anomaly localization , 2020, ICLR.

[43]  Bir Bhanu,et al.  Towards Visually Explaining Variational Autoencoders , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[44]  Hongxing He,et al.  Outlier Detection Using Replicator Neural Networks , 2002, DaWaK.

[45]  Thomas G. Dietterich,et al.  A Unifying Review of Deep and Shallow Anomaly Detection , 2020, Proceedings of the IEEE.

[46]  Carlos Guestrin,et al.  "Why Should I Trust You?": Explaining the Predictions of Any Classifier , 2016, ArXiv.

[47]  Klaus-Robert Müller,et al.  Towards Explaining Anomalies: A Deep Taylor Decomposition of One-Class Models , 2018, Pattern Recognit..

[48]  Randy C. Paffenroth,et al.  Anomaly Detection with Robust Deep Autoencoders , 2017, KDD.

[49]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[50]  Rethinking Assumptions in Deep Anomaly Detection , 2020, ArXiv.

[51]  Alexander Binder,et al.  Unmasking Clever Hans predictors and assessing what machines really learn , 2019, Nature Communications.

[52]  Robert P. W. Duin,et al.  Support Vector Data Description , 2004, Machine Learning.

[53]  Sanjay Chawla,et al.  Anomaly Detection using One-Class Neural Networks , 2018, ArXiv.

[54]  Alex Krizhevsky,et al.  Learning Multiple Layers of Features from Tiny Images , 2009 .

[55]  Wojciech Samek,et al.  Methods for interpreting and understanding deep neural networks , 2017, Digit. Signal Process..

[56]  Alexander Binder,et al.  Deep One-Class Classification , 2018, ICML.

[57]  Luc Van Gool,et al.  The Pascal Visual Object Classes (VOC) Challenge , 2010, International Journal of Computer Vision.

[58]  Roland Vollgraf,et al.  Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms , 2017, ArXiv.

[59]  Klaus-Robert Müller,et al.  Fairwashing Explanations with Off-Manifold Detergent , 2020, ICML.

[60]  Ran El-Yaniv,et al.  Deep Anomaly Detection Using Geometric Transformations , 2018, NeurIPS.

[61]  H. Arp.  Discordant observations. , 1990, Science.

[62]  Alexander Binder,et al.  Explaining nonlinear classification decisions with deep Taylor decomposition , 2015, Pattern Recognit..

[63]  Carlos Guestrin,et al.  Anchors: High-Precision Model-Agnostic Explanations , 2018, AAAI.

[64]  David M. J. Tax,et al.  One-class classification , 2001 .

[65]  Ankur Taly,et al.  Axiomatic Attribution for Deep Networks , 2017, ICML.

[66]  W. R. Buckland,et al.  Outliers in Statistical Data , 1979 .

[67]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.