Linear regression over networks with communication guarantees

Conference item

A key functionality of emerging connected autonomous systems such as smart cities, smart transportation systems, and the industrial Internet-of-Things, is the ability to process and learn from data collected at different physical locations. This is increasingly attracting attention under the terms of distributed learning and federated learning. However, in connected autonomous systems, data transfer takes place over communication networks with often limited resources. This paper examines algorithms for communication-efficient learning for linear regression tasks by exploiting the informativeness of the data. The developed algorithms enable a tradeoff between communication and learning with theoretical performance guarantees and efficient practical implementations.

Authors: Gatsis, K

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Journal of Machine Learning Research
• Host title: Proceedings of the Conference on Learning for Dynamics and Control
• Pages: 767-778
• Series: Proceedings of Machine Learning Research
• Series number: 144
• Publication date: 2021-05-29
• Acceptance date: 2021-03-06
• Event title: 3rd Annual Conference on Learning for Dynamics and Control Conference (L4DC 2021)
• Event location: Virtual event
• Event website: https://l4dc.ethz.ch/
• Event start date: 2021-06-07
• Event end date: 2021-06-08
• ISSN: 2640-3498
• Language: English
• Keywords: learning over networks; distributed learning; FFR; federated learning