Sample-free white-box out-of-distribution detection for deep learning

Being able to detect irrelevant test examples with respect to deployed deep learning models is paramount to properly and safely using them. In this paper, we address the problem of rejecting such out-of-distribution (OOD) samples in a fully sample-free way, i.e., without requiring any access to in- distribution or OOD samples. We propose several indicators which can be computed alongside the prediction with little additional cost, assuming white-box access to the network. These indicators prove useful, stable and complementary for OOD detection on frequently-used architectures. We also introduce a surprisingly simple, yet effective summary OOD indicator. This indicator is shown to perform well across several networks and datasets and can furthermore be easily tuned as soon as samples become available. Lastly, we discuss how to exploit this summary in real-world settings.

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

[2]  Hongxia Jin,et al.  Generalized ODIN: Detecting Out-of-Distribution Image Without Learning From Out-of-Distribution Data , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  Ran El-Yaniv,et al.  SelectiveNet: A Deep Neural Network with an Integrated Reject Option , 2019, ICML.

[4]  Kibok Lee,et al.  Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution Samples , 2017, ICLR.

[5]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

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

[7]  Graham W. Taylor,et al.  Learning Confidence for Out-of-Distribution Detection in Neural Networks , 2018, ArXiv.

[8]  Kilian Q. Weinberger,et al.  Densely Connected Convolutional Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[9]  Xia Zhu,et al.  Out-of-Distribution Detection Using an Ensemble of Self Supervised Leave-out Classifiers , 2018, ECCV.

[10]  Qi Tian,et al.  Data-Free Learning of Student Networks , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

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

[12]  Ronald Kemker,et al.  Are Out-of-Distribution Detection Methods Effective on Large-Scale Datasets? , 2019, ArXiv.

[13]  Aaron C. Courville,et al.  Detecting semantic anomalies , 2019, AAAI.

[14]  Rick Salay,et al.  Detecting Out-of-Distribution Inputs in Deep Neural Networks Using an Early-Layer Output , 2019, ArXiv.

[15]  Pushmeet Kohli,et al.  Contrastive Training for Improved Out-of-Distribution Detection , 2020, ArXiv.

[16]  Jonathon Shlens,et al.  Explaining and Harnessing Adversarial Examples , 2014, ICLR.

[17]  Stephan Günnemann,et al.  Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shift , 2018, NeurIPS.

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

[19]  Kurt Keutzer,et al.  ZeroQ: A Novel Zero Shot Quantization Framework , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

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

[21]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[22]  R. Venkatesh Babu,et al.  Zero-Shot Knowledge Distillation in Deep Networks , 2019, ICML.

[23]  R. Srikant,et al.  Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networks , 2017, ICLR.

[24]  Matthieu Cord,et al.  Addressing Failure Prediction by Learning Model Confidence , 2019, NeurIPS.

[25]  Luca Antiga,et al.  Automatic differentiation in PyTorch , 2017 .

[26]  Derek Hoiem,et al.  Dreaming to Distill: Data-Free Knowledge Transfer via DeepInversion , 2019, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[27]  Kibok Lee,et al.  A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks , 2018, NeurIPS.

[28]  James J. Little,et al.  Does Your Model Know the Digit 6 Is Not a Cat? A Less Biased Evaluation of "Outlier" Detectors , 2018, ArXiv.

[29]  Nikos Komodakis,et al.  Wide Residual Networks , 2016, BMVC.

[30]  Marin Orsic,et al.  Discriminative out-of-distribution detection for semantic segmentation , 2018, ArXiv.

[31]  Andrew Y. Ng,et al.  Reading Digits in Natural Images with Unsupervised Feature Learning , 2011 .

[32]  Chandramouli Shama Sastry,et al.  Detecting Out-of-Distribution Examples with In-distribution Examples and Gram Matrices , 2019, ArXiv.

[33]  Kevin Gimpel,et al.  A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks , 2016, ICLR.

[34]  Sebastian Nowozin,et al.  Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift , 2019, NeurIPS.

[35]  Jasper Snoek,et al.  Likelihood Ratios for Out-of-Distribution Detection , 2019, NeurIPS.

[36]  Kumar Sricharan,et al.  Building robust classifiers through generation of confident out of distribution examples , 2018, ArXiv.

[37]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[38]  Igor M. Quintanilha,et al.  Detecting Out-Of-Distribution Samples Using Low-Order Deep Features Statistics , 2018 .

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

[40]  Yinda Zhang,et al.  LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop , 2015, ArXiv.