Learning to Diversify for Single Domain Generalization

By Zijian Wang, Yadan Luo, Ruihong Qiu, Zi-Liang Huang, Mahsa Baktash. IEEE International Conference on Computer Vision, 2021

353 citations recorded by the source corpus.

Domain generalization (DG) aims to generalize a model trained on multiple source (i.e., training) domains to a distributionally different target (i.e., test) domain. In contrast to the conventional DG that strictly requires the availability of multiple source domains, this paper considers a more realistic yet challenging scenario, namely Single Domain Generalization (Single-DG), where only one source domain is available for training. In this scenario, the limited diversity may jeopardize the model generalization on unseen target domains. To tackle this problem, we propose a style-complement module to enhance the generalization power of the model by synthesizing images from diverse distributions that are complementary to the source ones. More specifically, we adopt a tractable upper bound of mutual information (MI) between the generated and source samples and perform a two-step optimization iteratively: (1) by minimizing the MI upper bound approximation for each sample pair, the generated images are forced to be diversified from the source samples; (2) subsequently, we maximize the MI between the samples from the same semantic category, which assists the network to learn discriminative features from diverse-styled images. Extensive experiments on three benchmark datasets demonstrate the superiority of our approach, which surpasses the state-of-the-art single-DG methods by up to 25.14%. The code will be publicly available at https://github.com/BUserName/Learning_to_diversify

Subfields: Artificial Intelligence, Cancer Research, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/1109f18a41ad21635ae30952be89f56f52295f39

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Zi-Liang Huang
2. Hongzhi Yin
3. Tao Qin
4. Zijian Wang
5. Ruihong Qiu
6. Jindong Wang
7. Yidong Ouyang
8. Chang Liu

In Cancer Research

1. Jindong Wang
2. Cuiling Lan

In Computer Vision and Pattern Recognition

1. Ruihong Qiu
2. Tao Qin
3. Zi-Liang Huang
4. Jindong Wang
5. Yidong Ouyang
6. Chang Liu
7. Hongzhi Yin
8. Cuiling Lan

Citations in the corpus

• Contrastive Learning for Representation Degeneration Problem in Sequential Recommendation (2021, background): “Contrastive learning has been widely used in various deep learning areas for its strong ability to help with the self-supervised learning [2, 4, 8, 10, 11, 13, 14, 16, 41, 45].”
• Single-Source Domain Expansion Network for Cross-Scene Hyperspectral Image Classification (2022, neutral): “alization) [42] and Style-Agnostic Network (SagNet) [43].”
• Generalizing to Unseen Domains: A Survey on Domain Generalization (2021, methodology): “Several methods designed different generation strategies [40], [52], [58], [59], [81], [100],”