Improving Transferability of Representations via Augmentation-Aware Self-Supervision
This work addresses a limitation in representation learning for computer vision, offering a simple, low-cost improvement to enhance transferability, though it is incremental as it builds on existing methods.
The paper tackles the problem that invariance to data augmentations in representation learning can harm downstream tasks sensitive to those augmentations, and proposes AugSelf, an auxiliary self-supervised loss that improves transferability by preserving augmentation-aware information, resulting in consistent gains across supervised and unsupervised methods in various transfer learning scenarios.
Recent unsupervised representation learning methods have shown to be effective in a range of vision tasks by learning representations invariant to data augmentations such as random cropping and color jittering. However, such invariance could be harmful to downstream tasks if they rely on the characteristics of the data augmentations, e.g., location- or color-sensitive. This is not an issue just for unsupervised learning; we found that this occurs even in supervised learning because it also learns to predict the same label for all augmented samples of an instance. To avoid such failures and obtain more generalizable representations, we suggest to optimize an auxiliary self-supervised loss, coined AugSelf, that learns the difference of augmentation parameters (e.g., cropping positions, color adjustment intensities) between two randomly augmented samples. Our intuition is that AugSelf encourages to preserve augmentation-aware information in learned representations, which could be beneficial for their transferability. Furthermore, AugSelf can easily be incorporated into recent state-of-the-art representation learning methods with a negligible additional training cost. Extensive experiments demonstrate that our simple idea consistently improves the transferability of representations learned by supervised and unsupervised methods in various transfer learning scenarios. The code is available at https://github.com/hankook/AugSelf.