Abstract:Distance metric learning (DML) is a critical factor for image analysis and pattern recognition. To learn a robust distance metric for a target task, we need abundant side information (i.e., the similarity/dissimilarity pairwise constraints over the labeled data), which is usually unavailable in practice due to the high labeling cost. This paper considers the transfer learning setting by exploiting the large quantity of side information from certain related, but different source tasks to help with target metric learning (with only a little side information). The state-of-the-art metric learning algorithms usually fail in this setting because the data distributions of the source task and target task are often quite different. We address this problem by assuming that the target distance metric lies in the space spanned by the eigenvectors of the source metrics (or other randomly generated bases). The target metric is represented as a combination of the base metrics, which are computed using the decomposed components of the source metrics (or simply a set of random bases); we call the proposed method, decomposition-based transfer DML (DTDML). In particular, DTDML learns a sparse combination of the base metrics to construct the target metric by forcing the target metric to be close to an integration of the source metrics. The main advantage of the proposed method compared with existing transfer metric learning approaches is that we directly learn the base metric coefficients instead of the target metric. To this end, far fewer variables need to be learned. We therefore obtain more reliable solutions given the limited side information and the optimization tends to be faster. Experiments on the popular handwritten image (digit, letter) classification and challenge natural image annotation tasks demonstrate the effectiveness of the proposed method.

Robust Metric Boosts Transfer

Multi-component transfer metric learning for handling unrelated source domain samples

Transfer Metric Learning: Algorithms, Applications and Outlooks.

Decomposition-based transfer distance metric learning for image classification.

Online Heterogeneous Transfer Metric Learning.

Toward Intelligent Product Retrieval for TV-to-Online (T2O) Application: A Transfer Metric Learning Approach

Heterogeneous Multitask Metric Learning Across Multiple Domains

General Heterogeneous Transfer Distance Metric Learning Via Knowledge Fragments Transfer.

Deep Model Transferability from Attribution Maps

Robust Transfer Learning Based on Minimax Principle

On Combining Side Information and Unlabeled Data for Heterogeneous Multi-Task Metric Learning.

Transferring Knowledge Fragments for Learning Distance Metric from A Heterogeneous Domain

Toward Effortless TV-to-Online (T2O) Experience: A Novel Metric Learning Approach.

NES-TL: Network Embedding Similarity-Based Transfer Learning

Enhancing the Accuracy of an Image Classification Model Using Cross-Modality Transfer Learning

Scalable Angular Discriminative Deep Metric Learning for Face Recognition

Deep Factorized Metric Learning

A Mathematical Framework for Quantifying Transferability in Multi-source Transfer Learning

Transferability-Guided Cross-Domain Cross-Task Transfer Learning

Two-Level Multi-Task Metric Learning with Application to Multi-Classification

Manifold Criterion Guided Transfer Learning via Intermediate Domain Generation