MixStyle Neural Networks for Domain Generalization and Adaptation.
Saved in:
| Title: | MixStyle Neural Networks for Domain Generalization and Adaptation. |
|---|---|
| Authors: | Zhou, Kaiyang1 (AUTHOR) kyzhou@hkbu.edu.hk, Yang, Yongxin2 (AUTHOR), Qiao, Yu3 (AUTHOR), Xiang, Tao4 (AUTHOR) |
| Source: | International Journal of Computer Vision. Mar2024, Vol. 132 Issue 3, p822-836. 15p. |
| Subjects: | Generalization, Stimulus generalization, Data augmentation, Image recognition (Computer vision), Reinforcement learning, Machine learning, Source code, Supervised learning |
| Abstract: | Neural networks do not generalize well to unseen data with domain shifts—a longstanding problem in machine learning and AI. To overcome the problem, we propose MixStyle, a simple plug-and-play, parameter-free module that can improve domain generalization performance without the need to collect more data or increase model capacity. The design of MixStyle is simple: it mixes the feature statistics of two random instances in a single forward pass during training. The idea is grounded by the finding from recent style transfer research that feature statistics capture image style information, which essentially defines visual domains. Therefore, mixing feature statistics can be seen as an efficient way to synthesize new domains in the feature space, thus achieving data augmentation. MixStyle is easy to implement with a few lines of code, does not require modification to training objectives, and can fit a variety of learning paradigms including supervised domain generalization, semi-supervised domain generalization, and unsupervised domain adaptation. Our experiments show that MixStyle can significantly boost out-of-distribution generalization performance across a wide range of tasks including image recognition, instance retrieval and reinforcement learning. The source code is released at https://github.com/KaiyangZhou/mixstyle-release. [ABSTRACT FROM AUTHOR] |
| Copyright of International Journal of Computer Vision is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Neural networks do not generalize well to unseen data with domain shifts—a longstanding problem in machine learning and AI. To overcome the problem, we propose MixStyle, a simple plug-and-play, parameter-free module that can improve domain generalization performance without the need to collect more data or increase model capacity. The design of MixStyle is simple: it mixes the feature statistics of two random instances in a single forward pass during training. The idea is grounded by the finding from recent style transfer research that feature statistics capture image style information, which essentially defines visual domains. Therefore, mixing feature statistics can be seen as an efficient way to synthesize new domains in the feature space, thus achieving data augmentation. MixStyle is easy to implement with a few lines of code, does not require modification to training objectives, and can fit a variety of learning paradigms including supervised domain generalization, semi-supervised domain generalization, and unsupervised domain adaptation. Our experiments show that MixStyle can significantly boost out-of-distribution generalization performance across a wide range of tasks including image recognition, instance retrieval and reinforcement learning. The source code is released at https://github.com/KaiyangZhou/mixstyle-release. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 09205691 |
| DOI: | 10.1007/s11263-023-01913-8 |
