Just DIAL: DomaIn Alignment Layers for Unsupervised Domain Adaptation
This addresses the challenge of visual recognition systems performing poorly in new settings without labeled target data, representing a novel method for a known bottleneck.
The paper tackles the problem of domain shift in unsupervised domain adaptation by introducing DomaIn Alignment Layers (DIAL) to match source and target data distributions to a reference distribution, achieving strong performance on three public benchmarks.
The empirical fact that classifiers, trained on given data collections, perform poorly when tested on data acquired in different settings is theoretically explained in domain adaptation through a shift among distributions of the source and target domains. Alleviating the domain shift problem, especially in the challenging setting where no labeled data are available for the target domain, is paramount for having visual recognition systems working in the wild. As the problem stems from a shift among distributions, intuitively one should try to align them. In the literature, this has resulted in a stream of works attempting to align the feature representations learned from the source and target domains. Here we take a different route. Rather than introducing regularization terms aiming to promote the alignment of the two representations, we act at the distribution level through the introduction of \emph{DomaIn Alignment Layers} (\DIAL), able to match the observed source and target data distributions to a reference one. Thorough experiments on three different public benchmarks we confirm the power of our approach.