Domain Adaptation via Maximizing Surrogate Mutual Information
This work addresses domain adaptation for machine learning applications where labeled data is scarce, but it appears incremental as it builds on existing mutual information approaches.
The paper tackles unsupervised domain adaptation by proposing SIDA, a framework that maximizes mutual information between features using a surrogate joint distribution, with experiments showing it is comparable to state-of-the-art methods on standard tasks.
Unsupervised domain adaptation (UDA) aims to predict unlabeled data from target domain with access to labeled data from the source domain. In this work, we propose a novel framework called SIDA (Surrogate Mutual Information Maximization Domain Adaptation) with strong theoretical guarantees. To be specific, SIDA implements adaptation by maximizing mutual information (MI) between features. In the framework, a surrogate joint distribution models the underlying joint distribution of the unlabeled target domain. Our theoretical analysis validates SIDA by bounding the expected risk on target domain with MI and surrogate distribution bias. Experiments show that our approach is comparable with state-of-the-art unsupervised adaptation methods on standard UDA tasks.