Search for: All records

We provide a compressive-measurement based method to detect susceptible agents who may receive misinformation through their contact with ‘stubborn agents’ whose goal is to influence the opinions of agents in the network. We consider a DeGroot-type opinion dynamics model where regular agents revise their opinions by linearly combining their neighbors’ opinions, but stubborn agents, while influencing others, do not change their opinions. Our proposed method hinges on estimating the temporal difference vector of network-wide opinions, computed at time instances when the stubborn agents interact. We show that this temporal difference vector has approximately the same support as the locations of the susceptible agents. Moreover, both the interaction instances and the temporal difference vector can be estimated from a small number of aggregated opinions. The performance of our method is studied both analytically and empirically. We show that the detection error decreases when the social network is better connected, or when the stubborn agents are ‘less talkative’. 

This paper considers the problem of inferring the topology of a graph from noisy outputs of an unknown graph filter excited by low-rank signals. Limited by this low-rank structure, we focus on the community detection problem, whose aim is to partition the node set of the unknown graph into subsets with high edge densities. We propose to detect the communities by applying spectral clustering on the low-rank output covariance matrix. To analyze the performance, we show that the low-rank covariance yields a sketch of the eigenvectors of the unknown graph. Importantly, we provide theoretical bounds on the error introduced by this sketching procedure based on spectral features of the graph filter involved. Finally, our theoretical findings are validated via numerical experiments in both synthetic and real-world graphs. 

This paper studies the security aspect of gossip-based decentralized optimization algorithms for multi agent systems against data injection attacks. Our contributions are two-fold. First, we show that the popular distributed projected gradient method (by Nedi´c et al.) can be attacked by coordinated insider attacks, in which the attackers are able to steer the final state to a point of their choosing. Second, we propose a metric that can be computed locally by the trustworthy agents processing their own iterates and those of their neighboring agents. This metric can be used by the trustworthy agents to detect and localize the attackers. We conclude the paper by supporting our findings with numerical experiments.