Distributed Generalized Linear Models: A Privacy-Preserving Approach
For practitioners needing privacy-preserving GLM training in distributed or federated environments, this offers a scalable solution, but the contribution appears incremental.
This paper extends a privacy-preserving linear regression approach to generalized linear models (GLMs), enabling computation in distributed and federated settings. Numerical studies show advantages over conventional maximum likelihood estimation, though no concrete performance numbers are provided.
This paper presents a novel approach to classical linear regression, enabling model computation from data streams or in a distributed setting while preserving data privacy in federated environments. We extend this framework to generalized linear models (GLMs), ensuring scalability and adaptability to diverse data distributions while maintaining privacy-preserving properties. To assess the effectiveness of our approach, we conduct numerical studies on both simulated and real datasets, comparing our method with conventional maximum likelihood estimation for GLMs using iteratively reweighted least squares. Our results demonstrate the advantages of the proposed method in distributed and federated settings.