AuthorsR. Money, J. Krishnan, B. Beferull-Lozano and E. Isufi
TitleScalable and Privacy-aware Online Learning of Nonlinear Structural Equation Models
AfilliationMachine Learning
Project(s)Signal and Information Processing for Intelligent Systems
StatusAccepted
Publication TypeJournal Article
Year of Publication2022
JournalIEEE Open Journal of Signal Processing
PublisherIEEE
Abstract

An online topology estimation algorithm for nonlinear structural equation models (SEM) is proposed in this paper, addressing the nonlinearity and the non-stationarity of real-world systems. The nonlinearity is modeled using kernel formulations, and the curse of dimensionality associated with the kernels is mitigated using random feature approximation. The online learning strategy uses a group-lasso-based optimization framework with a prediction-corrections technique that accounts for the model evolution. \E{The proposed approach has three properties of interest. First, it enjoys node-separable learning, which allows for scalability in large networks. Second, it offers privacy in SEM learning by replacing the actual data with node-specific random features. Third, its performance can be characterized theoretically via a dynamic regret analysis, showing that it is possible to obtain a linear dynamic regret bound under mild assumptions. Numerical results with synthetic and real data corroborate our findings and show competitive performance w.r.t. state-of-the-art alternatives.

Notes

This work is a joint collaboration between SimulaMet and University of Agder. This work was supported by the IKTPLUSS INDURB grant 270730/O70  from the Research Council of Norway.

DOI10.36227/techrxiv.21407952.v1
Citation Key42845