Multi-Step Decentralized Domain Adaptation

Akhil Mathur, Shaoduo Gan, Anton Isopoussu, Fahim Kawsar, Nadia Berthouze, Nicholas D. Lane

TL;DR: A novel method for decentralized and distributed domain adaptation, as a way to make these methods more practical in real ML systems.

Abstract: Despite the recent breakthroughs in unsupervised domain adaptation (uDA), no prior work has studied the challenges of applying these methods in practical machine learning scenarios. In this paper, we highlight two significant bottlenecks for uDA, namely excessive centralization and poor support for distributed domain datasets. Our proposed framework, MDDA, is powered by a novel collaborator selection algorithm and an effective distributed adversarial training method, and allows for uDA methods to work in a decentralized and privacy-preserving way.

Keywords: domain adaptation, decentralization