VAFL: a Method of Vertical Asynchronous Federated Learning
This addresses the problem of efficient and private vertical federated learning for applications like healthcare, though it appears incremental as it builds on existing FL concepts with new techniques.
The paper tackles vertical federated learning by developing an asynchronous method called VAFL that allows clients to run stochastic gradient algorithms independently, making it suitable for intermittent connectivity, and demonstrates favorable results compared to centralized and synchronous FL methods on image and healthcare datasets.
Horizontal Federated learning (FL) handles multi-client data that share the same set of features, and vertical FL trains a better predictor that combine all the features from different clients. This paper targets solving vertical FL in an asynchronous fashion, and develops a simple FL method. The new method allows each client to run stochastic gradient algorithms without coordination with other clients, so it is suitable for intermittent connectivity of clients. This method further uses a new technique of perturbed local embedding to ensure data privacy and improve communication efficiency. Theoretically, we present the convergence rate and privacy level of our method for strongly convex, nonconvex and even nonsmooth objectives separately. Empirically, we apply our method to FL on various image and healthcare datasets. The results compare favorably to centralized and synchronous FL methods.