Information-theoretic Semi-supervised Metric Learning via Entropy Regularization
This work addresses metric learning in semi-supervised settings, which is incremental as it builds on existing entropy regularization approaches.
The authors tackled the problem of semi-supervised metric learning without relying on the manifold assumption by proposing Seraph, which integrates labeled and unlabeled data through entropy regularization and low-rank projection, resulting in favorable performance compared to existing methods.
We propose a general information-theoretic approach called Seraph (SEmi-supervised metRic leArning Paradigm with Hyper-sparsity) for metric learning that does not rely upon the manifold assumption. Given the probability parameterized by a Mahalanobis distance, we maximize the entropy of that probability on labeled data and minimize it on unlabeled data following entropy regularization, which allows the supervised and unsupervised parts to be integrated in a natural and meaningful way. Furthermore, Seraph is regularized by encouraging a low-rank projection induced from the metric. The optimization of Seraph is solved efficiently and stably by an EM-like scheme with the analytical E-Step and convex M-Step. Experiments demonstrate that Seraph compares favorably with many well-known global and local metric learning methods.