Federated Variational Inference: Towards Improved Personalization and Generalization

Citations Over Time

Abstract

Conventional federated learning algorithms train a single global model by leveraging all participating clients’ data. However, due to heterogeneity in client generative distributions and predictive models, these approaches may not appropriately approximate the predictive process, converge to an optimal state, or generalize to new clients. We study personalization and generalization in stateless cross-device federated learning setups assuming heterogeneity in client data distributions and predictive models. We first propose a hierarchical generative model and formalize it using Bayesian Inference. We then approximate this process using Variational Inference to train our model efficiently. We call this algorithm Federated Variational Inference (FedVI). We use PAC-Bayes analysis to provide generalization bounds for FedVI. We evaluate our model on FEMNIST and CIFAR-100 image classification and show that FedVI beats the state-of-the-art on both tasks.

Related Papers

• → The Human Rights of Stateless Persons(2006)93 cited
• → Homegrown Statelessness in Malaysia: The Administratively Stateless and the Promise of the Principle of Genuine and Effective Links(2019)14 cited
• → Cookies and Session State(2003)
• → The concept and structure of status of stateless persons – international law approach(2023)