Information-Theoretical Learning of Discriminative Clusters for Unsupervised Domain Adaptation

TL;DR

This paper introduces a novel unsupervised domain adaptation method that jointly learns domain-invariant and discriminative features using an information-theoretic approach, improving classification accuracy without target labels.

Contribution

The paper proposes a new joint learning framework that combines domain invariance and discriminative feature learning via an information-theoretic metric, with effective gradient-based optimization.

Findings

Significant improvement over existing methods in object recognition and sentiment analysis.

Effective hyperparameter cross-validation without labeled target data.

Validated assumptions through empirical studies on benchmark tasks.

Abstract

We study the problem of unsupervised domain adaptation, which aims to adapt classifiers trained on a labeled source domain to an unlabeled target domain. Many existing approaches first learn domain-invariant features and then construct classifiers with them. We propose a novel approach that jointly learn the both. Specifically, while the method identifies a feature space where data in the source and the target domains are similarly distributed, it also learns the feature space discriminatively, optimizing an information-theoretic metric as an proxy to the expected misclassification error on the target domain. We show how this optimization can be effectively carried out with simple gradient-based methods and how hyperparameters can be cross-validated without demanding any labeled data from the target domain. Empirical studies on benchmark tasks of object recognition and sentiment analysis validated our modeling assumptions and demonstrated significant improvement of our method over competing ones in classification accuracies.