Unsupervised Domain Adaptation Using Approximate Label Matching

TL;DR

This paper introduces Approximate Label Matching (ALM), an unsupervised domain adaptation method that uses rough target labels to align source and target data, improving performance over existing techniques.

Contribution

ALM is a novel unsupervised domain adaptation approach that leverages noisy target labels to learn effective data transformations, outperforming traditional methods.

Findings

ALM outperforms common domain adaptation techniques on simulated data.

The learned transformation has favorable properties compared to other methods.

Regularization ensures the classifier cannot distinguish between source and transformed target samples.

Abstract

Domain adaptation addresses the problem created when training data is generated by a so-called source distribution, but test data is generated by a significantly different target distribution. In this work, we present approximate label matching (ALM), a new unsupervised domain adaptation technique that creates and leverages a rough labeling on the test samples, then uses these noisy labels to learn a transformation that aligns the source and target samples. We show that the transformation estimated by ALM has favorable properties compared to transformations estimated by other methods, which do not use any kind of target labeling. Our model is regularized by requiring that a classifier trained to discriminate source from transformed target samples cannot distinguish between the two. We experiment with ALM on simulated and real data, and show that it outperforms techniques commonly used in the field.