Interplay between Topology and Social Learning over Weak Graphs

TL;DR

This paper analyzes how network topology influences social learning and how the influence of sub-networks can be inferred from beliefs, revealing conditions for effective topology learning and leader-follower dynamics in weakly-connected graphs.

Contribution

It provides analytical formulas for beliefs in weak graphs, explores the interplay between social and topology learning, and identifies conditions enabling topology inference from belief data.

Findings

Beliefs depend on network topology and data heterogeneity.

Leader-follower behavior can emerge where some agents control others.

Topology learning is feasible when the number of hypotheses exceeds sending components.

Abstract

We consider a social learning problem, where a network of agents is interested in selecting one among a finite number of hypotheses. We focus on weakly-connected graphs where the network is partitioned into a sending part and a receiving part. The data collected by the agents might be heterogeneous. For example, some sub-networks might intentionally generate data from a fake hypothesis in order to influence other agents. The social learning task is accomplished via a diffusion strategy where each agent: i) updates individually its belief using its private data; ii) computes a new belief by exponentiating a linear combination of the log-beliefs of its neighbors. First, we examine what agents learn over weak graphs (social learning problem). We obtain analytical formulas for the beliefs at the different agents, which reveal how the agents' detection capability and the network topology interact to influence the beliefs. In particular, the formulas allow us to predict when a leader-follower behavior is possible, where some sending agents can control the mind of the receiving agents by forcing them to choose a particular hypothesis. Second, we consider the dual or reverse learning problem that reveals how agents learned: given a stream of beliefs collected at a receiving agent, we would like to discover the global influence that any sending component exerts on this receiving agent (topology learning problem). A remarkable and perhaps unexpected interplay between social and topology learning is observed: given $H$ hypotheses and $S$ sending components, topology learning can be feasible when $H\geq S$. The latter being only a necessary condition, we examine the feasibility of topology learning for two useful classes of problems. The analysis reveals that a critical element to enable faithful topology learning is the diversity in the statistical models of the sending sub-networks.