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.
If you find our work useful to your research, please cite:
@inproceedings{venkat2020msda,
title={Your Classifier can Secretly Suffice Multi-Source Domain Adaptation},
author={Venkat, Naveen and Kundu, Jogendra Nath and Singh, Durgesh Kumar and Revanur, Ambareesh and Babu, R. Venkatesh},
booktitle={Advances in Neural Information Processing Systems (NeurIPS)},
year={2020}
}
This project is licenced under the MIT Licence.
Please feel free to get in touch via email: nav.naveenvenkat@gmail.com