SALT: Subspace Alignment as an Auxiliary Learning Task for Domain Adaptation
Unsupervised domain adaptation aims to transfer and adapt knowledge learned from a labeled source domain to an unlabeled target domain. Key components of unsupervised domain adaptation include: (a) maximizing performance on the source, and (b) aligning the source and target domains. Traditionally, these tasks have either been considered as separate, or assumed to be implicitly addressed together with high-capacity feature extractors. In this paper, we advance a third broad approach; which we term SALT. The core idea is to consider alignment as an auxiliary task to the primary task of maximizing performance on the source. The auxiliary task is made rather simple by assuming a tractable data geometry in the form of subspaces. We synergistically allow certain parameters derived from the closed-form auxiliary solution, to be affected by gradients from the primary task. The proposed approach represents a unique fusion of geometric and model-based alignment with gradient-flows from a data-driven primary task. SALT is simple, rooted in theory, and outperforms state-of-the-art on multiple standard benchmarks.
NurtureToken New!

Token crowdsale for this paper ends in

Buy Nurture Tokens

Authors

Are you an author of this paper? Check the Twitter handle we have for you is correct.

Kowshik Thopalli (add twitter)
Jayaraman J. Thiagarajan (add twitter)
Rushil Anirudh (add twitter)
Pavan Turaga (add twitter)
Ask The Authors

Ask the authors of this paper a question or leave a comment.

Read it. Rate it.
#1. Which part of the paper did you read?

#2. The paper contains new data or analyses that is openly accessible?
#3. The conclusion is supported by the data and analyses?
#4. The conclusion is of scientific interest?
#5. The result is likely to lead to future research?

Github
User:
None (add)
Repo:
None (add)
Stargazers:
0
Forks:
0
Open Issues:
0
Network:
0
Subscribers:
0
Language:
None
Youtube
Link:
None (add)
Views:
0
Likes:
0
Dislikes:
0
Favorites:
0
Comments:
0
Other
Sample Sizes (N=):
Inserted:
Words Total:
Words Unique:
Source:
Abstract:
None
06/11/19 06:04PM
6,207
2,194
Tweets
kowshik0808: Our new work on unsupervised domain adaptation is now available on arxiv. SALT: Subspace Alignment as an Auxiliary Learning Task for Domain Adaptation -https://t.co/eml5FcEbaH Mentored by three amazing people. - @RUSH1L @jjayaram7 and @pturaga1 https://t.co/ihpyueQPqk
RUSH1L: New preprint: We find that decoupling domain alignment from the final task improves domain adaptation. A simple subspace based alignment consistently outperforms adversarial DA like CDAN etc. Exciting work from @kowshik0808, @jjayaram7 & @pturaga1 Paper: https://t.co/BV5S6BzuTQ https://t.co/vN7YCyioEa
StatsPapers: SALT: Subspace Alignment as an Auxiliary Learning Task for Domain Adaptation. https://t.co/Gn6EN6yR34
arxiv_cs_LG: SALT: Subspace Alignment as an Auxiliary Learning Task for Domain Adaptation. Kowshik Thopalli, Jayaraman J. Thiagarajan, Rushil Anirudh, and Pavan Turaga https://t.co/Mq3uBtY3Jm
BrundageBot: SALT: Subspace Alignment as an Auxiliary Learning Task for Domain Adaptation. Kowshik Thopalli, Jayaraman J. Thiagarajan, Rushil Anirudh, and Pavan Turaga https://t.co/X6hAho0Xet
Images
Related