Distributed networked learning with correlated data

• Published in: Automatica (Oxford) Vol. 137; p. 110134
• Main Authors: Hong, Lingzhou, Garcia, Alfredo, Eksin, Ceyhun
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.03.2022
• Subjects: Large scale optimization problems and methods; Learning theory; Multi-agent systems; Network-based computing systems; Parameter and state estimation; Statistical analysis; Large scale optimization problems and methods; Statistical analysis; Learning theory; Network-based computing systems; Parameter and state estimation; Multi-agent systems
• ISSN: 0005-1098; 1873-2836
• DOI: 10.1016/j.automatica.2021.110134

We consider a distributed estimation method in a setting with heterogeneous streams of correlated data distributed across nodes in a network. In the considered approach, linear models are estimated locally (i.e., with only local data) subject to a network regularization term that penalizes a local model that differs from neighboring models. We analyze computation dynamics (associated with stochastic gradient updates) and information exchange (associated with exchanging current models with neighboring nodes). We provide a finite-time characterization of convergence of the weighted ensemble average estimate and compare this result to federated learning, an alternative approach to estimation wherein a single model is updated by locally generated gradient updates. This comparison highlights the trade-off between speed vs precision: while model updates take place at a faster rate in federated learning, the proposed networked approach to estimation enables the identification of models with higher precision. We illustrate the method’s general applicability in two examples: estimating a Markov random field using wireless sensor networks and modeling prey escape behavior of flocking birds based on a publicly available dataset.