Unsupervised Domain Adaptation by Backpropagation

TL;DR

This paper introduces a simple yet effective method for unsupervised domain adaptation in deep neural networks, enabling models to learn domain-invariant features without labeled target data by using a gradient reversal layer.

Contribution

It proposes a novel gradient reversal layer that allows standard backpropagation to train models for domain adaptation without labeled target data.

Findings

Achieves significant domain adaptation performance on image classification tasks.

Outperforms previous state-of-the-art methods on Office datasets.

Works with minimal architectural modifications.

Abstract

Top-performing deep architectures are trained on massive amounts of labeled data. In the absence of labeled data for a certain task, domain adaptation often provides an attractive option given that labeled data of similar nature but from a different domain (e.g. synthetic images) are available. Here, we propose a new approach to domain adaptation in deep architectures that can be trained on large amount of labeled data from the source domain and large amount of unlabeled data from the target domain (no labeled target-domain data is necessary). As the training progresses, the approach promotes the emergence of "deep" features that are (i) discriminative for the main learning task on the source domain and (ii) invariant with respect to the shift between the domains. We show that this adaptation behaviour can be achieved in almost any feed-forward model by augmenting it with few standard layers and a simple new gradient reversal layer. The resulting augmented architecture can be trained using standard backpropagation. Overall, the approach can be implemented with little effort using any of the deep-learning packages. The method performs very well in a series of image classification experiments, achieving adaptation effect in the presence of big domain shifts and outperforming previous state-of-the-art on Office datasets.