New Book Chapter Published in the Research Handbook on Digital Data: "Collaborative distributed machine learning: a path to strengthen data privacy"

In this chapter, Sascha Rank, David Jin, Niclas Kannengießer, and Ali Sunyaev discuss Collaborative Distributed Machine Learning (CDML) as a paradigm that lets multiple parties jointly train ML models without sharing their raw data. The chapter introduces federated learning, gossip learning, and split learning, and shows how privacy-enhancing technologies, such as differential privacy, homomorphic encryption, secure multi-party computation, and trusted execution environments, can defend against attacks that would otherwise leak private information.

The chapter appears in the Research Handbook on Digital Data: Interdisciplinary Perspectives, which was edited by Aleksi Aaltonen, Marta Stelmaszak, and Kalle Lyytinen. The handbook brings together leading scholars across disciplines, including Geoffrey Bowker, Rob Kitchin, Ola Henfridsson, Sirkka Jarvenpaa, Jan Recker, Youngjin Yoo, and M. Lynne Markus.

You can find the chapter here: https://dx.doi.org/10.4337/9781035348718.00033