Anomalous Example Detection in Deep Learning: A Survey

Abstract

Deep Learning (DL) is vulnerable to out-of-distribution and adversarial examples resulting in incorrect outputs. To make DL more robust, several posthoc (or runtime) anomaly detection techniques to detect (and discard) these anomalous samples have been proposed in the recent past. This survey tries to provide a structured and comprehensive overview of the research on anomaly detection for DL based applications. We provide a taxonomy for existing techniques based on their underlying assumptions and adopted approaches. We discuss various techniques in each of the categories and provide the relative strengths and weaknesses of the approaches. Our goal in this survey is to provide an easier yet better understanding of the techniques belonging to different categories in which research has been done on this topic. Finally, we highlight the unsolved research challenges while applying anomaly detection techniques in DL systems and present some high-impact future research directions.

References in corpus (40)

• Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networks
• Wide Residual Networks
• A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks
• Deep Learning for Anomaly Detection: A Survey
• Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examples
• A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks
• Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift
• Domain Generalization via Invariant Feature Representation
• On Evaluating Adversarial Robustness
• Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networks
• Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionality
• EMNIST: an extension of MNIST to handwritten letters
• Deep Anomaly Detection with Outlier Exposure
• Deep Anomaly Detection Using Geometric Transformations
• Detecting Adversarial Samples from Artifacts
• On the (Statistical) Detection of Adversarial Examples
• To Trust Or Not To Trust A Classifier
• Understanding Measures of Uncertainty for Adversarial Example Detection
• WAIC, but Why? Generative Ensembles for Robust Anomaly Detection
• Adversarial Attacks on Deep Neural Networks for Time Series Classification
• Adversarial Attacks Against Medical Deep Learning Systems
• Characterizing Audio Adversarial Examples Using Temporal Dependency
• Improving Reconstruction Autoencoder Out-of-distribution Detection with Mahalanobis Distance
• RNN-based Early Cyber-Attack Detection for the Tennessee Eastman Process
• Characterizing Adversarial Examples Based on Spatial Consistency Information for Semantic Segmentation
• Open Category Detection with PAC Guarantees
• Identifying Unknown Unknowns in the Open World: Representations and Policies for Guided Exploration
• Probabilistic Modeling of Deep Features for Out-of-Distribution and Adversarial Detection
• Are Generative Classifiers More Robust to Adversarial Attacks?
• A Less Biased Evaluation of Out-of-distribution Sample Detectors
• HashTran-DNN: A Framework for Enhancing Robustness of Deep Neural Networks against Adversarial Malware Samples
• Defending Against Adversarial Examples with K-Nearest Neighbor
• Unsupervised Out-of-Distribution Detection with Batch Normalization
• Confidence from Invariance to Image Transformations
• Hyperparameter-Free Out-of-Distribution Detection Using Softmax of Scaled Cosine Similarity
• A Research Agenda: Dynamic Models to Defend Against Correlated Attacks
• Detecting Adversarial Examples and Other Misclassifications in Neural Networks by Introspection
• Detecting Adversarial Examples in Convolutional Neural Networks
• Analysis of Confident-Classifiers for Out-of-distribution Detection
• Attentioned Convolutional LSTM InpaintingNetwork for Anomaly Detection in Videos

Cited by in corpus (8)

• A Comprehensive Survey on Graph Anomaly Detection with Deep Learning
• Likelihood Regret: An Out-of-Distribution Detection Score For Variational Auto-encoder
• Counterfactual Zero-Shot and Open-Set Visual Recognition
• Embedded out-of-distribution detection on an autonomous robot platform
• Estimating the Brittleness of AI: Safety Integrity Levels and the Need for Testing Out-Of-Distribution Performance
• Anomaly Detection by Recombining Gated Unsupervised Experts
• A Survey on Assessing the Generalization Envelope of Deep Neural Networks: Predictive Uncertainty, Out-of-distribution and Adversarial Samples
• Joint Distribution across Representation Space for Out-of-Distribution Detection