Deep graph-level anomaly detection by glocal knowledge distillation

Graph-level anomaly detection (GAD) describes the problem of detecting graphs that are abnormal in their structure and/or the features of their nodes, as compared to other graphs. One of the challenges in GAD is to devise graph representations that enable the detection of both locally- and globally-...

Full description

Saved in:
Bibliographic Details
Main Authors: MA, Rongrong, PANG, Guansong, CHEN, Ling, HENGEL, Anton Van Den
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2022
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/7054
https://ink.library.smu.edu.sg/context/sis_research/article/8057/viewcontent/3488560.3498473.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
id sg-smu-ink.sis_research-8057
record_format dspace
spelling sg-smu-ink.sis_research-80572022-04-07T09:06:42Z Deep graph-level anomaly detection by glocal knowledge distillation MA, Rongrong PANG, Guansong CHEN, Ling HENGEL, Anton Van Den Graph-level anomaly detection (GAD) describes the problem of detecting graphs that are abnormal in their structure and/or the features of their nodes, as compared to other graphs. One of the challenges in GAD is to devise graph representations that enable the detection of both locally- and globally-anomalous graphs, i.e., graphs that are abnormal in their fine-grained (node-level) or holistic (graph-level) properties, respectively. To tackle this challenge we introduce a novel deep anomaly detection approach for GAD that learns rich global and local normal pattern information by joint random distillation of graph and node representations. The random distillation is achieved by training one GNN to predict another GNN with randomly initialized network weights. Extensive experiments on 16 real-world graph datasets from diverse domains show that our model significantly outperforms seven state-of-the-art models. Code and datasets are available at https://git.io/GLocalKD. 2022-02-01T08:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/7054 info:doi/10.1145/3488560.3498473 https://ink.library.smu.edu.sg/context/sis_research/article/8057/viewcontent/3488560.3498473.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Graph-level anomaly detection Graph neural networks Knowledge distillation Deep learning Artificial Intelligence and Robotics Graphics and Human Computer Interfaces
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic Graph-level anomaly detection
Graph neural networks
Knowledge distillation
Deep learning
Artificial Intelligence and Robotics
Graphics and Human Computer Interfaces
spellingShingle Graph-level anomaly detection
Graph neural networks
Knowledge distillation
Deep learning
Artificial Intelligence and Robotics
Graphics and Human Computer Interfaces
MA, Rongrong
PANG, Guansong
CHEN, Ling
HENGEL, Anton Van Den
Deep graph-level anomaly detection by glocal knowledge distillation
description Graph-level anomaly detection (GAD) describes the problem of detecting graphs that are abnormal in their structure and/or the features of their nodes, as compared to other graphs. One of the challenges in GAD is to devise graph representations that enable the detection of both locally- and globally-anomalous graphs, i.e., graphs that are abnormal in their fine-grained (node-level) or holistic (graph-level) properties, respectively. To tackle this challenge we introduce a novel deep anomaly detection approach for GAD that learns rich global and local normal pattern information by joint random distillation of graph and node representations. The random distillation is achieved by training one GNN to predict another GNN with randomly initialized network weights. Extensive experiments on 16 real-world graph datasets from diverse domains show that our model significantly outperforms seven state-of-the-art models. Code and datasets are available at https://git.io/GLocalKD.
format text
author MA, Rongrong
PANG, Guansong
CHEN, Ling
HENGEL, Anton Van Den
author_facet MA, Rongrong
PANG, Guansong
CHEN, Ling
HENGEL, Anton Van Den
author_sort MA, Rongrong
title Deep graph-level anomaly detection by glocal knowledge distillation
title_short Deep graph-level anomaly detection by glocal knowledge distillation
title_full Deep graph-level anomaly detection by glocal knowledge distillation
title_fullStr Deep graph-level anomaly detection by glocal knowledge distillation
title_full_unstemmed Deep graph-level anomaly detection by glocal knowledge distillation
title_sort deep graph-level anomaly detection by glocal knowledge distillation
publisher Institutional Knowledge at Singapore Management University
publishDate 2022
url https://ink.library.smu.edu.sg/sis_research/7054
https://ink.library.smu.edu.sg/context/sis_research/article/8057/viewcontent/3488560.3498473.pdf
_version_ 1770576205259472896