ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Your classifier can secretly suffice multi-source domain adaptation

Venkat, N and Nath, J and Durgesh, K and Singh, K and Revanur, A and Babu, RV (2020) Your classifier can secretly suffice multi-source domain adaptation. In: 34th Conference on Neural Information Processing Systems, NeurIPS 2020, 6-12 December, 2020, virtual.

[img] PDF
NeurIPS-2020.pdf - Published Version
Restricted to Registered users only

Download (1MB) | Request a copy


Multi-Source Domain Adaptation (MSDA) deals with the transfer of task knowledge from multiple labeled source domains to an unlabeled target domain, under a domain-shift. Existing methods aim to minimize this domain-shift using auxiliary distribution alignment objectives. In this work, we present a different perspective to MSDA wherein deep models are observed to implicitly align the domains under label supervision. Thus, we aim to utilize implicit alignment without additional training objectives to perform adaptation. To this end, we use pseudo-labeled target samples and enforce a classifier agreement on the pseudo-labels, a process called Self-supervised Implicit Alignment (SImpAl). We find that SImpAl readily works even under category-shift among the source domains. Further, we propose classifier agreement as a cue to determine the training convergence, resulting in a simple training algorithm. We provide a thorough evaluation of our approach on five benchmarks, along with detailed insights into each component of our approach. © 2020 Neural information processing systems foundation. All rights reserved.

Item Type: Conference Paper
Publication: Advances in Neural Information Processing Systems
Publisher: Neural information processing systems foundation
Additional Information: The copyright for this article belongs to Neural information processing systems foundation
Keywords: Classifier agreements; Domain adaptation; Multi-Sources; Target domain; Task knowledge; Training algorithms, Alignment
Department/Centre: Division of Interdisciplinary Sciences > Computational and Data Sciences
Date Deposited: 03 Aug 2021 09:28
Last Modified: 04 Aug 2021 11:54
URI: http://eprints.iisc.ac.in/id/eprint/69040

Actions (login required)

View Item View Item