Transfer Learning for Latent Variable Network Models
Akhil Jalan, Arya Mazumdar, Soumendu Sundar Mukherjee, Purnamrita Sarkar
TL;DR
The paper addresses transfer learning for latent-variable network models, where the edge probability matrix $Q$ is learned using a fully observed source $P$ and sparse target data. It introduces a rankings-based algorithm that exploits shared latent structure via a graph-distance to align latent positions without requiring a parametric form, achieving vanishing estimation error. The authors establish a minimax lower bound for Stochastic Block Models and show a simple clustering-based method attains the rate in the appropriate regime, supported by empirical results on simulated and real networks that illustrate effective cross-domain transfer. Overall, the work demonstrates practical transfer capabilities for estimating latent-network structures across domains, with theoretically justified rates and real-data validation.
Abstract
We study transfer learning for estimation in latent variable network models. In our setting, the conditional edge probability matrices given the latent variables are represented by $P$ for the source and $Q$ for the target. We wish to estimate $Q$ given two kinds of data: (1) edge data from a subgraph induced by an $o(1)$ fraction of the nodes of $Q$, and (2) edge data from all of $P$. If the source $P$ has no relation to the target $Q$, the estimation error must be $Ω(1)$. However, we show that if the latent variables are shared, then vanishing error is possible. We give an efficient algorithm that utilizes the ordering of a suitably defined graph distance. Our algorithm achieves $o(1)$ error and does not assume a parametric form on the source or target networks. Next, for the specific case of Stochastic Block Models we prove a minimax lower bound and show that a simple algorithm achieves this rate. Finally, we empirically demonstrate our algorithm's use on real-world and simulated graph transfer problems.