Heterogeneous Domain Adaptation Through Progressive Alignment

• Published in: IEEE transaction on neural networks and learning systems Vol. 30; no. 5; pp. 1381 - 1391
• Main Authors: Li, Jingjing, Lu, Ke, Huang, Zi, Zhu, Lei, Shen, Heng Tao
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.05.2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Adaptation; Alignment; Cross-modal; Dictionary learning; Divergence; domain adaptation; Encoding; feature representation; Image classification; Image recognition; Learning systems; Machine learning; Object recognition; Objective function; Optimization; Task analysis; Text categorization; Text recognition; Transfer learning
• ISSN: 2162-237X; 2162-2388
• DOI: 10.1109/TNNLS.2018.2868854

In real-world transfer learning tasks, especially in cross-modal applications, the source domain and the target domain often have different features and distributions, which are well known as the heterogeneous domain adaptation (HDA) problem. Yet, existing HDA methods focus on either alleviating the feature discrepancy or mitigating the distribution divergence due to the challenges of HDA. In fact, optimizing one of them can reinforce the other. In this paper, we propose a novel HDA method that can optimize both feature discrepancy and distribution divergence in a unified objective function. Specifically, we present progressive alignment , which first learns a new transferable feature space by dictionary-sharing coding, and then aligns the distribution gaps on the new space. Different from previous HDA methods that are limited to specific scenarios, our approach can handle diverse features with arbitrary dimensions. Extensive experiments on various transfer learning tasks, such as image classification, text categorization, and text-to-image recognition, verify the superiority of our method against several state-of-the-art approaches.