Domain Generalization Under Conditional and Label Shifts via Variational Bayesian Inference
Abstract
In this work, we propose a domain generalization (DG) approach to learn on several labeled source domains and transfer knowledge to a target domain that is inaccessible in training. Considering the inherent conditional and label shifts, we would expect the alignment of p(x|y) and p(y). However, the widely used domain invariant feature learning (IFL) methods relies on aligning the marginal concept shift w.r.t. p(x), which rests on an unrealistic assumption that p(y) is invariant across domains. We thereby propose a novel variational Bayesian inference framework to enforce the conditional distribution alignment w.r.t. p(x|y) via the prior distribution matching in a latent space, which also takes the marginal label shift w.r.t. p(y) into consideration with the posterior alignment. Extensive experiments on various benchmarks demonstrate that our framework is robust to the label shift and the cross-domain accuracy is significantly improved, thereby achieving superior performance over the conventional IFL counterparts.
Cite
Text
Liu et al. "Domain Generalization Under Conditional and Label Shifts via Variational Bayesian Inference." International Joint Conference on Artificial Intelligence, 2021. doi:10.24963/IJCAI.2021/122Markdown
[Liu et al. "Domain Generalization Under Conditional and Label Shifts via Variational Bayesian Inference." International Joint Conference on Artificial Intelligence, 2021.](https://mlanthology.org/ijcai/2021/liu2021ijcai-domain/) doi:10.24963/IJCAI.2021/122BibTeX
@inproceedings{liu2021ijcai-domain,
title = {{Domain Generalization Under Conditional and Label Shifts via Variational Bayesian Inference}},
author = {Liu, Xiaofeng and Hu, Bo and Jin, Linghao and Han, Xu and Xing, Fangxu and Ouyang, Jinsong and Lu, Jun and El Fakhri, Georges and Woo, Jonghye},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2021},
pages = {881-887},
doi = {10.24963/IJCAI.2021/122},
url = {https://mlanthology.org/ijcai/2021/liu2021ijcai-domain/}
}