FRuDA: Framework for Distributed Adversarial Domain Adaptation

TL;DR

FRuDA is a novel framework enabling effective and efficient distributed adversarial unsupervised domain adaptation across multiple devices, addressing the limitations of centralized algorithms in real-world distributed environments.

Contribution

It introduces FRuDA, the first end-to-end distributed adversarial uDA framework, with two new algorithms that enhance accuracy and training efficiency in distributed settings.

Findings

Boosts target domain accuracy by up to 50%.

Improves training efficiency by at least 11 times.

Validates effectiveness on five image and speech datasets.

Abstract

Breakthroughs in unsupervised domain adaptation (uDA) can help in adapting models from a label-rich source domain to unlabeled target domains. Despite these advancements, there is a lack of research on how uDA algorithms, particularly those based on adversarial learning, can work in distributed settings. In real-world applications, target domains are often distributed across thousands of devices, and existing adversarial uDA algorithms -- which are centralized in nature -- cannot be applied in these settings. To solve this important problem, we introduce FRuDA: an end-to-end framework for distributed adversarial uDA. Through a careful analysis of the uDA literature, we identify the design goals for a distributed uDA system and propose two novel algorithms to increase adaptation accuracy and training efficiency of adversarial uDA in distributed settings. Our evaluation of FRuDA with five image and speech datasets show that it can boost target domain accuracy by up to 50% and improve the training efficiency of adversarial uDA by at least 11 times.