Metric Learning across Heterogeneous Domains by Respectively Aligning Both Priors and Posteriors
This addresses the challenge of domain adaptation in machine learning, particularly for scenarios with limited labeled data in the target domain, though it appears incremental as it builds on existing metric learning and alignment concepts.
The paper tackles the problem of learning a single metric across two heterogeneous domains with different data availability, where the source domain has many labeled samples and the target domain has few labeled but many unlabeled samples. The result is a model that aligns priors and posteriors in a common space, validated on cross-language retrieval and cross-domain object recognition tasks.
In this paper, we attempts to learn a single metric across two heterogeneous domains where source domain is fully labeled and has many samples while target domain has only a few labeled samples but abundant unlabeled samples. To the best of our knowledge, this task is seldom touched. The proposed learning model has a simple underlying motivation: all the samples in both the source and the target domains are mapped into a common space, where both their priors P(sample)s and their posteriors P(label|sample)s are forced to be respectively aligned as much as possible. We show that the two mappings, from both the source domain and the target domain to the common space, can be reparameterized into a single positive semi-definite(PSD) matrix. Then we develop an efficient Bregman Projection algorithm to optimize the PDS matrix over which a LogDet function is used to regularize. Furthermore, we also show that this model can be easily kernelized and verify its effectiveness in crosslanguage retrieval task and cross-domain object recognition task.