Advances and Open Problems in Federated Learning

Other free copies

• Centre National de la Recherche Scientifique (hal.inria.fr)Preprint · Web page
• Centre National de la Recherche Scientifique (inria.hal.science)Preprint · Web page
• École Polytechnique Fédérale de Lausanne (infoscience.epfl.ch)Preprint · Web page
• Centre d'Études en Civilisations, Langues et Lettres Étrangères (hdl.handle.net)Preprint · Web page
• Centre d'Études en Civilisations, Langues et Lettres Étrangères (hdl.handle.net)Preprint · Web page
• Cornell University (doi.org)Free copy · Web page

Abstract

Federated learning (FL) is a machine learning setting where many clients (e.g., mobile devices or whole organizations) collaboratively train a model under the orchestration of a central server (e.g., service provider), while keeping the training data decentralized. FL embodies the principles of focused data collection and minimization, and can mitigate many of the systemic privacy risks and costs resulting from traditional, centralized machine learning and data science approaches. Motivated by the explosive growth in FL research, this monograph discusses recent advances and presents an extensive collection of open problems and challenges.