CUDA: Contradistinguisher for unsupervised domain adaptation

Balgi, S and Dukkipati, A (2019) CUDA: Contradistinguisher for unsupervised domain adaptation. In: 19th IEEE International Conference on Data Mining, ICDM 2019, 8-11 November, 2019, Beijing; China, pp. 21-30.

PDF ICDM.pdf - Published Version Restricted to Registered users only Download (1MB) | Request a copy

Abstract

Humans are very sophisticated in learning new information on a completely unknown domain because humans can contradistinguish, i.e., distinguish by contrasting qualities. We learn on a new unknown domain by jointly using unsupervised information directly from unknown domain and supervised information previously acquired knowledge from some other domain. Motivated by this supervised-unsupervised joint learning, we propose a simple model referred as Contradistinguisher (CTDR) for unsupervised domain adaptation whose objective is to jointly learn to contradistinguish on unlabeled target domain in a fully unsupervised manner along with prior knowledge acquired by supervised learning on an entirely different domain. Most recent works in domain adaptation rely on an indirect way of first aligning the source and target domain distributions and then learn a classifier on labeled source domain to classify target domain. This approach of indirect way of addressing the real task of unlabeled target domain classification has three main drawbacks. (i) The sub-task of obtaining a perfect alignment of the domain in itself might be impossible due to large domain shift (e.g., language domains). (ii) The use of multiple classifiers to align the distributions, unnecessarily increases the complexity of the neural networks leading to over-fitting in many cases. (iii) Due to distribution alignment, the domain specific information is lost as the domains get morphed. In this work, we propose a simple and direct approach that does not require domain alignment. We jointly learn CTDR on both source and target distribution for unsupervised domain adaptation task using contradistinguish loss for the unlabeled target domain in conjunction with supervised loss for labeled source domain. Our experiments show that avoiding domain alignment by directly addressing the task of unlabeled target domain classification using CTDR achieves state-of-the-art results on eight visual and four language benchmark domain adaptation datasets.

Item Type:	Conference Paper
Publication:	Proceedings - IEEE International Conference on Data Mining, ICDM
Publisher:	Institute of Electrical and Electronics Engineers Inc.
Additional Information:	cited By 0; Conference of 19th IEEE International Conference on Data Mining, ICDM 2019 ; Conference Date: 8 November 2019 Through 11 November 2019; Conference Code:157223
Keywords:	Alignment; Computer vision; Data mining; Deep learning; Machine learning; Sentiment analysis; Unsupervised learning; Visual languages, Benchmark domains; Different domains; Domain adaptation; Domain-specific information; Feature learning; Multiple classifiers; State of the art; Transfer learning, Classification (of information)
Department/Centre:	Division of Electrical Sciences > Computer Science & Automation
Date Deposited:	11 Jun 2020 10:46
Last Modified:	11 Jun 2020 10:46
URI:	http://eprints.iisc.ac.in/id/eprint/64610

Actions (login required)

View Item