Deep Learning for Anomaly Detection: A Review

Abstract

Anomaly detection, a.k.a. outlier detection or novelty detection, has been a lasting yet active research area in various research communities for several decades. There are still some unique problem complexities and challenges that require advanced approaches. In recent years, deep learning enabled anomaly detection, i.e., deep anomaly detection, has emerged as a critical direction. This paper surveys the research of deep anomaly detection with a comprehensive taxonomy, covering advancements in three high-level categories and 11 fine-grained categories of the methods. We review their key intuitions, objective functions, underlying assumptions, advantages and disadvantages, and discuss how they address the aforementioned challenges. We further discuss a set of possible future opportunities and new perspectives on addressing the challenges.

Survey paper, 36 pages, 180 references, 2 figures, 4 tables

References in corpus (2)

• Coupling Learning of Complex Interactions
• VEWS: A Wikipedia Vandal Early Warning System

Cited by in corpus (112)

• Deep Neural Networks and Tabular Data: A Survey
• A Comprehensive Survey on Graph Anomaly Detection with Deep Learning
• Deep Learning for Time Series Anomaly Detection: A Survey
• Deep Isolation Forest for Anomaly Detection
• Anomaly Detection on Attributed Networks via Contrastive Self-Supervised Learning
• ECOD: Unsupervised Outlier Detection Using Empirical Cumulative Distribution Functions
• Machine Learning in Aerodynamic Shape Optimization
• An Evaluation of Anomaly Detection and Diagnosis in Multivariate Time Series
• Deep Learning for Unsupervised Anomaly Localization in Industrial Images: A Survey
• Feature Encoding with AutoEncoders for Weakly-supervised Anomaly Detection
• Deep Learning in Deterministic Computational Mechanics
• Data-driven Machinery Fault Diagnosis: A Comprehensive Review
• CARLA: Self-supervised Contrastive Representation Learning for Time Series Anomaly Detection
• A Survey on Unsupervised Anomaly Detection Algorithms for Industrial Images
• DTAAD: Dual Tcn-Attention Networks for Anomaly Detection in Multivariate Time Series Data
• Autoencoders for unsupervised anomaly detection in high energy physics
• Toward Deep Supervised Anomaly Detection: Reinforcement Learning from Partially Labeled Anomaly Data
• Self-Supervised Anomaly Detection in Computer Vision and Beyond: A Survey and Outlook
• Deep Graph-level Anomaly Detection by Glocal Knowledge Distillation
• Predicting the Future of AI with AI: High-quality link prediction in an exponentially growing knowledge network
• Astronomia ex machina: a history, primer, and outlook on neural networks in astronomy
• Anomaly Detection in Dynamic Graphs: A Comprehensive Survey
• DASVDD: Deep Autoencoding Support Vector Data Descriptor for Anomaly Detection
• SiFall: Practical Online Fall Detection with RF Sensing
• Is it worth it? Comparing six deep and classical methods for unsupervised anomaly detection in time series
• Semi-supervised detection of structural damage using Variational Autoencoder and a One-Class Support Vector Machine
• Anomaly Detection in Aerial Videos with Transformers
• A Hybrid Deep Learning Anomaly Detection Framework for Intrusion Detection
• Adaptive Model Pooling for Online Deep Anomaly Detection from a Complex Evolving Data Stream
• Hybrid Classical-Quantum Autoencoder for Anomaly Detection
• Improving Generalizability of Graph Anomaly Detection Models via Data Augmentation
• Detecting Outliers with Foreign Patch Interpolation
• Unsupervised Outlier Detection using Memory and Contrastive Learning
• Lifelong Continual Learning for Anomaly Detection: New Challenges, Perspectives, and Insights
• Domain knowledge-informed Synthetic fault sample generation with Health Data Map for cross-domain Planetary Gearbox Fault Diagnosis
• REB: Reducing Biases in Representation for Industrial Anomaly Detection
• The Clever Hans Effect in Unsupervised Learning
• Fast Deep Autoencoder for Federated learning
• Meta-survey on outlier and anomaly detection
• Entropic Out-of-Distribution Detection: Seamless Detection of Unknown Examples
• PseudoBound: Limiting the anomaly reconstruction capability of one-class classifiers using pseudo anomalies
• Fuzzy Granule Density-Based Outlier Detection with Multi-Scale Granular Balls
• Multiple Instance Learning for Cheating Detection and Localization in Online Examinations
• CARE to Compare: A real-world dataset for anomaly detection in wind turbine data
• A new measure between sets of probability distributions with applications to erratic financial behavior
• A method to challenge symmetries in data with self-supervised learning
• Explainable Contextual Anomaly Detection using Quantile Regression Forests
• Active anomaly detection based on deep one-class classification
• Attention2Minority: A salient instance inference-based multiple instance learning for classifying small lesions in whole slide images
• Anomaly Detection Based on Isolation Mechanisms: A Survey
• MAD: Self-Supervised Masked Anomaly Detection Task for Multivariate Time Series
• Lifelong Learning on Evolving Graphs Under the Constraints of Imbalanced Classes and New Classes
• Detecting ergodic bubbles at the crossover to many-body localization using neural networks
• Quantum Transfer Learning with Adversarial Robustness for Classification of High-Resolution Image Datasets
• Few-shot Weakly-supervised Cybersecurity Anomaly Detection
• A benchmark for computational analysis of animal behavior, using animal-borne tags
• Quantum Generative Learning for High-Resolution Medical Image Generation
• ALGAN: Anomaly Detection by Generating Pseudo Anomalous Data via Latent Variables
• Unsupervised Graph Outlier Detection: Problem Revisit, New Insight, and Superior Method
• Anomaly detection and community detection in networks
• Explainable AI guided unsupervised fault diagnostics for high-voltage circuit breakers
• GRAM: An Interpretable Approach for Graph Anomaly Detection using Gradient Attention Maps
• Analysis of the first Genetic Engineering Attribution Challenge
• Black-box Error Diagnosis in Deep Neural Networks for Computer Vision: a Survey of Tools
• Convolutional Neural Network Design and Evaluation for Real-Time Multivariate Time Series Fault Detection in Spacecraft Attitude Sensors
• Lessons from the Development of an Anomaly Detection Interface on the Mars Perseverance Rover using the ISHMAP Framework
• SoftED: Metrics for Soft Evaluation of Time Series Event Detection
• Bidirectional skip-frame prediction for video anomaly detection with intra-domain disparity-driven attention
• Lightning Fast Video Anomaly Detection via Adversarial Knowledge Distillation
• Towards Phytoplankton Parasite Detection Using Autoencoders
• GI-PIP: Do We Require Impractical Auxiliary Dataset for Gradient Inversion Attacks?
• OneFlow: One-class flow for anomaly detection based on a minimal volume region
• Angel or Devil: Discriminating Hard Samples and Anomaly Contaminations for Unsupervised Time Series Anomaly Detection
• Double-Adversarial Activation Anomaly Detection: Adversarial Autoencoders are Anomaly Generators
• Object classification on video data of meteors and meteor-like phenomena: algorithm and data
• Quadratic Neuron-empowered Heterogeneous Autoencoder for Unsupervised Anomaly Detection
• Little Help Makes a Big Difference: Leveraging Active Learning to Improve Unsupervised Time Series Anomaly Detection
• Optimal Transport for Efficient, Unsupervised Anomaly Detection on Industrial Data
• PAC-Wrap: Semi-Supervised PAC Anomaly Detection
• A real-time anomaly detection method for robots based on a flexible and sparse latent space
• Diagnostic Spatio-temporal Transformer with Faithful Encoding
• Towards an Awareness of Time Series Anomaly Detection Models' Adversarial Vulnerability
• An Unsupervised Machine Learning Approach to Identify Spectral Energy Distribution Outliers: Application to the S-PLUS DR4 data
• SincVAE: A new semi-supervised approach to improve anomaly detection on EEG data using SincNet and variational autoencoder
• SPINEX: Similarity-based Predictions with Explainable Neighbors Exploration for Anomaly and Outlier Detection
• TriP-LLM: A Tri-Branch Patch-wise Large Language Model Framework for Time-Series Anomaly Detection
• Learning novel representations of variable sources from multi-modal $\text{Gaia}$ data via autoencoders
• EntropyStop: Unsupervised Deep Outlier Detection with Loss Entropy
• R2-AD2: Detecting Anomalies by Analysing the Raw Gradient
• Deep Learning for Prawn Farming: Forecasting and Anomaly Detection
• Detecting Abnormal Operations in Concentrated Solar Power Plants from Irregular Sequences of Thermal Images
• Appearance Blur-driven AutoEncoder and Motion-guided Memory Module for Video Anomaly Detection
• Comparative Study on Semi-supervised Learning Applied for Anomaly Detection in Hydraulic Condition Monitoring System
• Adaptive Anomaly Detection in Network Flows with Low-Rank Tensor Decompositions and Deep Unrolling
• Anomaly Detection by Recombining Gated Unsupervised Experts
• Detecting Compromised IoT Devices Using Autoencoders with Sequential Hypothesis Testing
• Unsupervised Learning for Industrial Defect Detection: A Case Study on Shearographic Data
• Automated all-sky detection of γ Doradus / δ Scuti hybrids in TESS data from positive unlabelled (PU) learning
• Temporal Convolution-based Hybrid Model Approach with Representation Learning for Real-Time Acoustic Anomaly Detection
• OrionBench: Benchmarking Time Series Generative Models in the Service of the End-User
• Quantum Patch-Based Autoencoder for Anomaly Segmentation
• Critical Review for One-class Classification: recent advances and the reality behind them
• Can I trust my anomaly detection system? A case study based on explainable AI
• Federated Attention Autoencoders with a Stochastic Aggregation Scheme for Anomaly Detection
• Mobiprox: Supporting Dynamic Approximate Computing on Mobiles
• Exploring exotic configurations with anomalous features using deep learning: Application of classical and quantum-classical hybrid anomaly detection
• Circuit design in biology and machine learning. II. Anomaly detection
• ESN-DAGMM: A Lightweight Framework for Unsupervised Time-Series Data Monitoring in 5G O-RAN Networks
• A Deep Learning Approach to Multipath Component Detection in Power Delay Profiles
• On Using the Shapley Value for Anomaly Localization: A Statistical Investigation
• Fast Unsupervised Deep Outlier Model Selection with Hypernetworks
• Wind turbine condition monitoring based on intra- and inter-farm federated learning