Information-theoretic regularization for Multi-source Domain Adaptation

• Published in: 2021 IEEE/CVF International Conference on Computer Vision (ICCV) pp. 9194 - 9203
• Main Authors: Park, Geon Yeong, Wan Lee, Sang
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.10.2021
• Subjects: Adversarial learning; Computational modeling; Computer architecture; Computer network reliability; Computer vision; Image synthesis; Scalability; Training; Transfer/Low-shot/Semi/Unsupervised Learning
• ISSN: 2380-7504
• DOI: 10.1109/ICCV48922.2021.00908

Adversarial learning strategy has demonstrated remarkable performance in dealing with single-source Domain Adaptation (DA) problems, and it has recently been applied to Multi-source DA (MDA) problems. Although most existing MDA strategies rely on a multiple domain discriminator setting, its effect on the latent space representations has been poorly understood. Here we adopt an information-theoretic approach to identify and. resolve the potential adverse effect of the multiple domain discriminators on MDA: disintegration of domain-discriminative information, limited computational scalability, and a large variance in the gradient of the loss during training. We examine the above issues by situating adversarial DA in the context of information regularization. This also provides a theoretical justification for using a single and unified domain discriminator. Based on this idea, we implement a novel neural architecture called a Multi-source Information-regularized Adaptation Networks (MIAN). Large-scale experiments demonstrate that MIAN, despite its structural simplicity, reliably and significantly outperforms other state-of-the-art methods.