Learning Graph Structures With Transformer for Multivariate Time-Series Anomaly Detection in IoT
Обучение структур графов с помощью Transformer для обнаружения аномалий в многомерных временных рядах в Интернете вещей
2021-07-27
SCID: 54.1/w7jzc8ns
Discuss with AI
Gumbel-softmax graph learningInfluence Propagation convolutionIoT sensor networksTransformer-based architecturemultivariate time series anomaly detection
Figures from the paper
Abstract (AI)
Many real-world Internet of Things (IoT) systems, which include a variety of Internet-connected sensory devices, produce substantial amounts of multivariate time-series data. Meanwhile, vital IoT infrastructures, such as smart power grids and water distribution networks are frequently targeted by cyberattacks, making anomaly detection an important study topic. Modeling such relatedness is, nevertheless, unavoidable for any efficient and effective anomaly detection system, given the intricate topological and nonlinear connections that are originally unknown among sensors. Furthermore, detecting anomalies in multivariate time series is difficult due to their temporal dependency and stochasticity. This article presented GTA, a new framework for multivariate time-series anomaly detection that involves automatically learning a graph structure, graph convolution, and modeling temporal dependency using a transformer-based architecture. The connection learning policy, which is based on the Gumbel-softmax sampling approach to learn bidirected links among sensors directly, is at the heart of the learning graph structure. To describe the anomaly information flow between network nodes, we introduced a new graph convolution called influence propagation convolution. In addition, to tackle the quadratic complexity barrier, we suggested a multibranch attention mechanism to replace the original multihead self-attention method. Extensive experiments on four publicly available anomaly detection benchmarks further demonstrate the superiority of our approach over alternative state of the arts.
Key Findings
1
A multi-branch attention mechanism replaces standard multi-head self-attention to address its quadratic computational complexity.
2
Experiments on four public anomaly-detection benchmarks show that GTA outperforms alternative state-of-the-art methods.
3
GTA is a framework for multivariate IoT time-series anomaly detection that jointly learns sensor graphs, performs graph convolution, and models temporal dependencies with a Transformer-based architecture.
4
Influence Propagation convolution is introduced to model anomaly information flow between network nodes.
5
The framework automatically learns directed sensor connections using Gumbel-softmax sampling, avoiding reliance on known topology or predefined relationships.
Research Object
multivariate time series generated by IoT sensor networks and infrastructures
Research Subject
anomalies and their spatiotemporal dependencies and inter-sensor influence relationships for detection
Publication Details
Publication Date
2021-07-27
Journal
Publisher
ISSN
Cited by
642
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai10
Generative adversarial networks2020
Graph neural networks: A review of methods and applications2020
Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting2021
Dynamic Graph CNN for Learning on Point Clouds2019
Robust Anomaly Detection for Multivariate Time Series through Stochastic Recurrent Neural Network2019
A Review on Outlier/Anomaly Detection in Time Series Data2021
Deep Learning for IoT Big Data and Streaming Analytics: A Survey2018
Graph Neural Network-Based Anomaly Detection in Multivariate Time Series2021
Multivariate Time-Series Anomaly Detection via Graph Attention Network2020
DeepAnT: A Deep Learning Approach for Unsupervised Anomaly Detection in Time Series2018
Cited by8
Transformers in Time Series: A Survey2023
Deep Learning for Time Series Anomaly Detection: A Survey2024
Deep Learning for Anomaly Detection in Time-Series Data: Review, Analysis, and Guidelines2021
A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detection2024
DCdetector: Dual Attention Contrastive Representation Learning for Time Series Anomaly Detection2023
Interactive AI With Retrieval-Augmented Generation for Next Generation Networking2024
Anomaly Detection in Dynamic Graphs: A Comprehensive Survey2024
Time Series Anomaly Detection Leveraging MSE Feedback with AutoEncoder and RNN2024