Anomaly Detection with Generative Adversarial Networks for Multivariate Time Series

Abstract

Today's Cyber-Physical Systems (CPSs) are large, complex, and affixed with networked sensors and actuators that are targets for cyber-attacks. Conventional detection techniques are unable to deal with the increasingly dynamic and complex nature of the CPSs. On the other hand, the networked sensors and actuators generate large amounts of data streams that can be continuously monitored for intrusion events. Unsupervised machine learning techniques can be used to model the system behaviour and classify deviant behaviours as possible attacks. In this work, we proposed a novel Generative Adversarial Networks-based Anomaly Detection (GAN-AD) method for such complex networked CPSs. We used LSTM-RNN in our GAN to capture the distribution of the multivariate time series of the sensors and actuators under normal working conditions of a CPS. Instead of treating each sensor's and actuator's time series independently, we model the time series of multiple sensors and actuators in the CPS concurrently to take into account of potential latent interactions between them. To exploit both the generator and the discriminator of our GAN, we deployed the GAN-trained discriminator together with the residuals between generator-reconstructed data and the actual samples to detect possible anomalies in the complex CPS. We used our GAN-AD to distinguish abnormal attacked situations from normal working conditions for a complex six-stage Secure Water Treatment (SWaT) system. Experimental results showed that the proposed strategy is effective in identifying anomalies caused by various attacks with high detection rate and low false positive rate as compared to existing methods.

arXiv commentThis paper was presented in the 7th International Workshop on Big Data, Streams and Heterogeneous Source Mining: Algorithms, Systems, Programming Models and Applications on the ACM Knowledge Discovery and Data Mining conference, August 2018, London, United Kingdom

References in corpus (7)

• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• SegAN: Adversarial Network with Multi-scale $L_1$ Loss for Medical Image Segmentation
• Efficient GAN-Based Anomaly Detection
• C-RNN-GAN: Continuous recurrent neural networks with adversarial training
• Real-valued (Medical) Time Series Generation with Recurrent Conditional GANs
• Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery
• A Survey of Classification Techniques in the Area of Big Data

Cited by in corpus (43)

• Deep Learning for Anomaly Detection: A Survey
• A Novel GAN-based Fault Diagnosis Approach for Imbalanced Industrial Time Series
• At the Dawn of Generative AI Era: A Tutorial-cum-Survey on New Frontiers in 6G Wireless Intelligence
• Probabilistic Forecasting of Sensory Data with Generative Adversarial Networks - ForGAN
• Likelihood Regret: An Out-of-Distribution Detection Score For Variational Auto-encoder
• Machine Learning in NextG Networks via Generative Adversarial Networks
• TCGAN: Convolutional Generative Adversarial Network for Time Series Classification and Clustering
• A Spatiotemporal Deep Neural Network for Fine-Grained Multi-Horizon Wind Prediction
• Modern Machine Learning Tools for Monitoring and Control of Industrial Processes: A Survey
• Negative Selection Algorithm Research and Applications in the last decade: A Review
• Multivariate Time-series Anomaly Detection via Graph Attention Network
• MTS-DVGAN: Anomaly Detection in Cyber-Physical Systems using a Dual Variational Generative Adversarial Network
• Unsupervised Deep Anomaly Detection for Multi-Sensor Time-Series Signals
• Learning Graph Structures with Transformer for Multivariate Time Series Anomaly Detection in IoT
• Unsupervised Out-of-Distribution Detection with Batch Normalization
• Exathlon: A Benchmark for Explainable Anomaly Detection over Time Series
• The Hidden Uncertainty in a Neural Networks Activations
• Evaluation of Point Pattern Features for Anomaly Detection of Defect within Random Finite Set Framework
• ESAD: End-to-end Deep Semi-supervised Anomaly Detection
• OneFlow: One-class flow for anomaly detection based on a minimal volume region
• Out-of-Distribution Detection Using Neural Rendering Generative Models
• DROCC: Deep Robust One-Class Classification
• Clustering with Outlier Removal
• Improving Robustness on Seasonality-Heavy Multivariate Time Series Anomaly Detection
• Hypergraph Learning based Recommender System for Anomaly Detection, Control and Optimization
• IXmon: Detecting and Analyzing DRDoS Attacks at Internet Exchange Points
• PAGAN: Portfolio Analysis with Generative Adversarial Networks
• Self-Supervised Representation Learning for Visual Anomaly Detection
• Enhanced Isotropy Maximization Loss: Seamless and High-Performance Out-of-Distribution Detection Simply Replacing the SoftMax Loss
• Anomaly Detection in Astronomical Images with Generative Adversarial Networks
• Enhancing Unsupervised Anomaly Detection with Score-Guided Network
• PSEUDo: Interactive Pattern Search in Multivariate Time Series with Locality-Sensitive Hashing and Relevance Feedback
• Masked Contrastive Learning for Anomaly Detection
• Acoustic anomaly detection via latent regularized gaussian mixture generative adversarial networks
• CVAD: A generic medical anomaly detector based on Cascade VAE
• Event Cause Analysis in Distribution Networks using Synchro Waveform Measurements
• Learning who is in the market from time series: market participant discovery through adversarial calibration of multi-agent simulators
• MIM-Based GAN: Information Metric to Amplify Small Probability Events Importance in Generative Adversarial Networks
• Multi-Class Anomaly Detection
• IPOF: An Extremely and Excitingly Simple Outlier Detection Booster via Infinite Propagation
• Deep Unsupervised Image Anomaly Detection: An Information Theoretic Framework
• Deep Clustering based Fair Outlier Detection
• Deep Neural Network based Wide-Area Event Classification in Power Systems