Mixture Data for Training Cannot Ensure Out-of-distribution Generalization

no code implementations • 25 Dec 2023 • Songming Zhang, Yuxiao Luo, Qizhou Wang, Haoang Chi, Xiaofeng Chen, Bo Han, Jinyan Li

Deep neural networks often face generalization problems to handle out-of-distribution (OOD) data, and there remains a notable theoretical gap between the contributing factors and their respective impacts.

Data Augmentation Out-of-Distribution Generalization