Improving out-of-distribution generalization via multi-task self-supervised pretraining
Abstract
Self-supervised feature representations have been shown to be useful for supervised classification, few-shot learning, and adversarial robustness. We show that features obtained using self-supervised learning are comparable to, or better than, supervised learning for domain generalization in computer vision. We introduce a new self-supervised pretext task of predicting responses to Gabor filter banks and demonstrate that multi-task learning of compatible pretext tasks improves domain generalization performance as compared to training individual tasks alone. Features learnt through self-supervision obtain better generalization to unseen domains when compared to their supervised counterpart when there is a larger domain shift between training and test distributions and even show better localization ability for objects of interest. Self-supervised feature representations can also be combined with other domain generalization methods to further boost performance.
Cite
@article{arxiv.2003.13525,
title = {Improving out-of-distribution generalization via multi-task self-supervised pretraining},
author = {Isabela Albuquerque and Nikhil Naik and Junnan Li and Nitish Keskar and Richard Socher},
journal= {arXiv preprint arXiv:2003.13525},
year = {2020}
}