ReAct: Out-of-distribution Detection With Rectified Activations

Out-of-distribution (OOD) detection has received much attention lately due to its practical importance in enhancing the safe deployment of neural networks. One of the primary challenges is that models often produce highly confident predictions on OOD data, which undermines the driving principle in OOD detection that the model should only be confident about in-distribution samples. In this work, we propose ReAct—a simple and effective technique for reducing model overconfidence on OOD data. Our method is motivated by novel analysis on internal activations of neural networks, which displays highly distinctive signature patterns for OOD distributions. Our method can generalize effectively to different network architectures and different OOD detection scores. We empirically demonstrate that ReAct achieves competitive detection performance on a comprehensive suite of benchmark datasets, and give theoretical explication for our method’s efficacy. On the ImageNet benchmark, ReAct reduces the false positive rate (FPR95) by 25.05% compared to the previous best method1.

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

[2]  Max Welling,et al.  Auto-Encoding Variational Bayes , 2013, ICLR.

[3]  A. Krizhevsky.  Convolutional Deep Belief Networks on CIFAR-10 , 2010 .

[4]  Alex Graves,et al.  Conditional Image Generation with PixelCNN Decoders , 2016, NIPS.

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

[6]  Jordi Luque,et al.  Input complexity and out-of-distribution detection with likelihood-based generative models , 2020, ICLR.

[7]  Samy Bengio,et al.  Density estimation using Real NVP , 2016, ICLR.

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

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

[10]  Avanti Shrikumar,et al.  Calibration with Bias-Corrected Temperature Scaling Improves Domain Adaptation Under Label Shift in Modern Neural Networks , 2019, ArXiv.

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

[12]  Alexander J. Smola,et al.  Detecting and Correcting for Label Shift with Black Box Predictors , 2018, ICML.

[13]  Pingmei Xu,et al.  TurkerGaze: Crowdsourcing Saliency with Webcam based Eye Tracking , 2015, ArXiv.

[14]  C. K. Chow,et al.  On optimum recognition error and reject tradeoff , 1970, IEEE Trans. Inf. Theory.

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

[16]  Yee Whye Teh,et al.  Do Deep Generative Models Know What They Don't Know? , 2018, ICLR.

[17]  Rui Huang,et al.  MOS: Towards Scaling Out-of-distribution Detection for Large Semantic Space , 2021, 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[18]  Kamyar Azizzadenesheli,et al.  Regularized Learning for Domain Adaptation under Label Shifts , 2019, ICLR.

[19]  Krista A. Ehinger,et al.  SUN database: Large-scale scene recognition from abbey to zoo , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[20]  Yoshua Bengio,et al.  How transferable are features in deep neural networks? , 2014, NIPS.

[21]  Iasonas Kokkinos,et al.  Describing Textures in the Wild , 2013, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[22]  Yang Song,et al.  The iNaturalist Species Classification and Detection Dataset , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[23]  Matthew Botvinick,et al.  On the importance of single directions for generalization , 2018, ICLR.

[24]  Alexei A. Efros,et al.  Test-Time Training with Self-Supervision for Generalization under Distribution Shifts , 2019, ICML.

[25]  Bolei Zhou,et al.  Revisiting the Importance of Individual Units in CNNs via Ablation , 2018, ArXiv.

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

[27]  Jason Yosinski,et al.  Deep neural networks are easily fooled: High confidence predictions for unrecognizable images , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

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

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

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

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

[32]  Weitang Liu,et al.  Energy-based Out-of-distribution Detection , 2020, NeurIPS.

[33]  Thomas G. Dietterich,et al.  Benchmarking Neural Network Robustness to Common Corruptions and Perturbations , 2018, ICLR.

[34]  Terrance E. Boult,et al.  Towards Open Set Deep Networks , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[35]  Jure Leskovec,et al.  WILDS: A Benchmark of in-the-Wild Distribution Shifts , 2021, ICML.

[36]  Rob Fergus,et al.  Visualizing and Understanding Convolutional Networks , 2013, ECCV.

[37]  Yixuan Li,et al.  Generalized Out-of-Distribution Detection: A Survey , 2021, ArXiv.

[38]  Yali Amit,et al.  Likelihood Regret: An Out-of-Distribution Detection Score For Variational Auto-encoder , 2020, NeurIPS.

[39]  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).

[40]  Yixuan Li,et al.  Can multi-label classification networks know what they don't know? , 2021, ArXiv.

[41]  Daan Wierstra,et al.  Stochastic Backpropagation and Approximate Inference in Deep Generative Models , 2014, ICML.

[42]  John E. Hopcroft,et al.  Stacked Generative Adversarial Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[43]  Fabio Roli,et al.  Support Vector Machines with Embedded Reject Option , 2002, SVM.

[44]  Marco Saerens,et al.  Adjusting the Outputs of a Classifier to New a Priori Probabilities: A Simple Procedure , 2002, Neural Computation.

[45]  Bin Dai,et al.  Further Analysis of Outlier Detection with Deep Generative Models , 2020, NeurIPS.

[46]  Kaiming He,et al.  Group Normalization , 2018, ECCV.

[47]  Kilian Q. Weinberger,et al.  Online Adaptation to Label Distribution Shift , 2021, NeurIPS.

[48]  Bolei Zhou,et al.  Places: A 10 Million Image Database for Scene Recognition , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[49]  Charles Blundell,et al.  Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles , 2016, NIPS.

[50]  Jian Sun,et al.  Identity Mappings in Deep Residual Networks , 2016, ECCV.

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

[52]  Hod Lipson,et al.  Convergent Learning: Do different neural networks learn the same representations? , 2015, FE@NIPS.

[53]  Andrew Gordon Wilson,et al.  Why Normalizing Flows Fail to Detect Out-of-Distribution Data , 2020, NeurIPS.

[54]  Mark Sandler,et al.  MobileNetV2: Inverted Residuals and Linear Bottlenecks , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[55]  G. S. Mudholkar,et al.  The epsilon-skew-normal distribution for analyzing near-normal data , 2000 .

[56]  Yixuan Li,et al.  On the Importance of Gradients for Detecting Distributional Shifts in the Wild , 2021, NeurIPS.

[57]  Sergey Levine,et al.  Can Autonomous Vehicles Identify, Recover From, and Adapt to Distribution Shifts? , 2020, ICML.

[58]  E. Tabak,et al.  A Family of Nonparametric Density Estimation Algorithms , 2013 .

[59]  Tim Salimans,et al.  Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks , 2016, NIPS.

[60]  Avanti Shrikumar,et al.  Maximum Likelihood with Bias-Corrected Calibration is Hard-To-Beat at Label Shift Adaptation , 2020, ICML.

[61]  Tonio Ball,et al.  Understanding Anomaly Detection with Deep Invertible Networks through Hierarchies of Distributions and Features , 2020, NeurIPS.

[62]  Matthias Hein,et al.  Why ReLU Networks Yield High-Confidence Predictions Far Away From the Training Data and How to Mitigate the Problem , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).