Diversity Boosted Learning for Domain Generalization with a Large Number of Domains

XI LENG, Xiaoying Tang, Yatao Bian

Keywords: Domain Generalization, Spurious Correlation

TL;DR: We propose a novel sampling framework to efficiently sample the most informative domains and data points to help train robust models against two kinds of spurious correlations in Domain Generalization field.

Abstract: Machine learning algorithms minimizing the average training loss typically suffer from poor generalization performance. It inspires various works for domain generalization (DG), among which a series of methods work by $O(n^2)$ pairwise domain operations with n domains, where each one is often costly. Moreover, while a common objective in the DG literature is to learn invariant representations against spurious correlations induced by domains, we point out the insufficiency of it and highlight the importance of alleviating spurious correlations caused by objects. Based on the observation that diversity helps mitigate spurious correlations, we propose a Diversity boosted twO-level saMplIng framework (DOMI) to efficiently sample the most informative ones among a large number of domains and data points. We show that DOMI helps train robust models against spurious correlations from both domain-side and object-side, substantially enhancing the performance of five backbone DG algorithms on Rotated MNIST and Rotated Fashion MNIST.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Deep Learning and representational learning