Consistent Tomography under Partial Observations over Adaptive Networks

TL;DR

This paper investigates the possibility of inferring agent influence in adaptive networks with partial observations, demonstrating that under certain conditions, the network's interaction profile can be consistently reconstructed as the network grows.

Contribution

It establishes that for symmetric combination policies, the influence structure can be reliably identified using clustering, even with limited and indirect observations.

Findings

Interacting and non-interacting pairs form separate clusters as network size increases.

Clustering algorithms can identify influencing agents under the studied conditions.

Results extend to asymmetric policies relevant for causation analysis.

Abstract

This work studies the problem of inferring whether an agent is directly influenced by another agent over an adaptive diffusion network. Agent i influences agent j if they are connected (according to the network topology), and if agent j uses the data from agent i to update its online statistic. The solution of this inference task is challenging for two main reasons. First, only the output of the diffusion learning algorithm is available to the external observer that must perform the inference based on these indirect measurements. Second, only output measurements from a fraction of the network agents is available, with the total number of agents itself being also unknown. The main focus of this article is ascertaining under these demanding conditions whether consistent tomography is possible, namely, whether it is possible to reconstruct the interaction profile of the observable portion of the network, with negligible error as the network size increases. We establish a critical achievability result, namely, that for symmetric combination policies and for any given fraction of observable agents, the interacting and non-interacting agent pairs split into two separate clusters as the network size increases. This remarkable property then enables the application of clustering algorithms to identify the interacting agents influencing the observations. We provide a set of numerical experiments that verify the results for finite network sizes and time horizons. The numerical experiments show that the results hold for asymmetric combination policies as well, which is particularly relevant in the context of causation.