Transfer Metric Learning for Unseen Domains

• Published in: Data Science and Engineering Vol. 5; no. 2; pp. 140 - 151
• Main Authors: Kumagai, Atsutoshi, Iwata, Tomoharu, Fujiwara, Yasuhiro
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.06.2020; Springer; Springer Nature B.V; SpringerOpen
• Subjects: Algorithm Analysis and Problem Complexity; Artificial Intelligence; Artificial neural networks; Chemistry and Earth Sciences; Clustering; Computer Science; Data Mining and Knowledge Discovery; Data points; Database Management; Domain generalization; Embedding; Labels; Learning; Lower bounds; Metric learning; Neural networks; Physics; Statistics for Engineering; Systems and Data Security; Training; Transfer learning; Transfer learning; Metric learning; Domain generalization
• ISSN: 2364-1185; 2364-1541
• DOI: 10.1007/s41019-020-00125-1

We propose a transfer metric learning method to infer domain-specific data embeddings for unseen domains, from which no data are given in the training phase, by using knowledge transferred from related domains. When training and test distributions are different, the standard metric learning cannot infer appropriate data embeddings. The proposed method can infer appropriate data embeddings for the unseen domains by using latent domain vectors, which are latent representations of domains and control the property of data embeddings for each domain. This latent domain vector is inferred by using a neural network that takes the set of feature vectors in the domain as an input. The neural network is trained without the unseen domains. The proposed method can instantly infer data embeddings for the unseen domains without (re)-training once the sets of feature vectors in the domains are given. To accumulate knowledge in advance, the proposed method uses labeled and unlabeled data in multiple source domains. Labeled data, i.e., data with label information such as class labels or pair (similar/dissimilar) constraints, are used for learning data embeddings in such a way that similar data points are close and dissimilar data points are separated in the embedding space. Although unlabeled data do not have labels, they have geometric information that characterizes domains. The proposed method incorporates this information in a natural way on the basis of a probabilistic framework. The conditional distributions of the latent domain vectors, the embedded data, and the observed data are parameterized by neural networks and are optimized by maximizing the variational lower bound using stochastic gradient descent. The effectiveness of the proposed method was demonstrated through experiments using three clustering tasks.