Unveiling Anomalous Edges and Nominal Connectivity of Attributed Networks

  title={Unveiling Anomalous Edges and Nominal Connectivity of Attributed Networks},
  author={Konstantinos D. Polyzos and Costas Mavromatis and Vassilis N. Ioannidis and Georgios B. Giannakis},
  journal={2020 54th Asilomar Conference on Signals, Systems, and Computers},
Uncovering anomalies in attributed networks has recently gained popularity due to its importance in unveiling outliers and flagging adversarial behavior in a gamut of data and network science applications including the Internet of Things (IoT), finance, security, to list a few. The present work deals with uncovering anomalous edges in attributed graphs using two distinct formulations with complementary strengths, which can be easily distributed, and hence efficient. The first relies on… 
5 Citations

Figures and Tables from this paper

Ensemble Gaussian Processes for Online Learning Over Graphs With Adaptivity and Scalability

This work capitalizes on Gaussian processes (GPs) to offer a Bayesian SSL approach over graphs with uncertainty quantification, a key attribute especially in safety-critical domains, to allow learning with scalability and preserve privacy.

Online Graph-Guided Inference Using Ensemble Gaussian Processes of Egonet Features

A Bayesian interval learning framework that utilizes Gaussian processes (GPs) to allow for uncertainty quantification – a key component in safety-critical applications.

Active Sampling over Graphs for Bayesian Reconstruction with Gaussian Ensembles

A suite of novel and intuitive AFs are developed to guide the active sampling process that leverages an adaptive Bayesian model comprising an ensemble of Gaussian Processes with enhanced expressiveness of the function space.

Graph-Adaptive Incremental Learning Using an Ensemble of Gaussian Process Experts

This work tackles the graph-guided semi-supervised learning task in a Gaussian process (GP) based Bayesian framework to propagate the distribution of nonparametric function estimates to allow for uncertainty quantification in safety-critical applications.

Robust Deep Learning on Graphs

This dissertation presents a meta-modelling system that automates the very labor-intensive and therefore time-heavy and therefore expensive and expensive process of systematically cataloging and cataloging individual components of a system.



Signal and Graph Perturbations via Total Least-Squares

This work introduces regularized total least-squares (TLS) based approaches and corresponding alternating minimization algorithms with convergence guarantees and tests on simulated data corroborate the effectiveness of the novel TLS-based approaches.

Graph-Based Learning Under Perturbations via Total Least-Squares

This work introduces a regularized total least-squares (TLS) approach and iterative algorithms with convergence guarantees to solve both graph-based learning tasks topology identification and inference of signals over graphs.

Joint Community and Anomaly Tracking in Dynamic Networks

A novel approach for jointly tracking communities while detecting such anomalous nodes in time-varying networks is advocated by postulating edge creation as the result of mutual community participation by node pairs, and a dynamic factor model with anomalous memberships captured through a sparse outlier matrix is put forth.

Non-Negative Residual Matrix Factorization with Application to Graph Anomaly Detection

NrMF, a non-negative residual matrix factorization framework, is proposed to address challenges given an IP source-destination traffic network and can naturally capture abnormal behaviors on graphs.

Dynamic Anomalography: Tracking Network Anomalies Via Sparsity and Low Rank

Comprehensive numerical tests with both synthetic and real network data corroborate the effectiveness of the proposed online algorithms and their tracking capabilities, and demonstrate that they outperform state-of-the-art approaches developed to diagnose traffic anomalies.

Kernel-based Inference of Functions over Graphs

Topology Identification and Learning over Graphs: Accounting for Nonlinearities and Dynamics

The main goal of this paper is to outline overarching advances, and develop a principled framework to capture nonlinearities through kernels, which are judiciously chosen from a preselected dictionary to optimally fit the data.

Defending Graph Convolutional Networks Against Adversarial Attacks

  • V. N. IoannidisG. Giannakis
  • Computer Science
    ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2020
This paper introduces graph neural network architectures that are robust to perturbed networked data that utilizes a randomization layer that performs link-dithering (LD) by adding or removing links with probabilities selected to boost robustness.

oddball: Spotting Anomalies in Weighted Graphs

Several new rules in density, weights, ranks and eigenvalues that seem to govern the so-called “neighborhood sub-graphs” are discovered and shown how to use these rules for anomaly detection.

TensorSplat: Spotting Latent Anomalies in Time

The method TENSORSPLAT, at the heart of which lies the "PARAFAC" decomposition method, can give good insights about the large networks that are of interest nowadays, and contributes to spotting micro-clusters, changes and, in general, anomalies.