Taxonomy-Structured Domain Adaptation

Tianyi Liu, Zihao Xu, Hao He, Guang-Yuan Hao, Guang-He Lee, Hao Wang
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:22215-22232, 2023.

Abstract

Domain adaptation aims to mitigate distribution shifts among different domains. However, traditional formulations are mostly limited to categorical domains, greatly simplifying nuanced domain relationships in the real world. In this work, we tackle a generalization with taxonomy-structured domains, which formalizes domains with nested, hierarchical similarity structures such as animal species and product catalogs. We build on the classic adversarial framework and introduce a novel taxonomist, which competes with the adversarial discriminator to preserve the taxonomy information. The equilibrium recovers the classic adversarial domain adaptation’s solution if given a non-informative domain taxonomy (e.g., a flat taxonomy where all leaf nodes connect to the root node) while yielding non-trivial results with other taxonomies. Empirically, our method achieves state-of-the-art performance on both synthetic and real-world datasets with successful adaptation.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-liu23ap, title = {Taxonomy-Structured Domain Adaptation}, author = {Liu, Tianyi and Xu, Zihao and He, Hao and Hao, Guang-Yuan and Lee, Guang-He and Wang, Hao}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {22215--22232}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/liu23ap/liu23ap.pdf}, url = {https://proceedings.mlr.press/v202/liu23ap.html}, abstract = {Domain adaptation aims to mitigate distribution shifts among different domains. However, traditional formulations are mostly limited to categorical domains, greatly simplifying nuanced domain relationships in the real world. In this work, we tackle a generalization with taxonomy-structured domains, which formalizes domains with nested, hierarchical similarity structures such as animal species and product catalogs. We build on the classic adversarial framework and introduce a novel taxonomist, which competes with the adversarial discriminator to preserve the taxonomy information. The equilibrium recovers the classic adversarial domain adaptation’s solution if given a non-informative domain taxonomy (e.g., a flat taxonomy where all leaf nodes connect to the root node) while yielding non-trivial results with other taxonomies. Empirically, our method achieves state-of-the-art performance on both synthetic and real-world datasets with successful adaptation.} }
Endnote
%0 Conference Paper %T Taxonomy-Structured Domain Adaptation %A Tianyi Liu %A Zihao Xu %A Hao He %A Guang-Yuan Hao %A Guang-He Lee %A Hao Wang %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-liu23ap %I PMLR %P 22215--22232 %U https://proceedings.mlr.press/v202/liu23ap.html %V 202 %X Domain adaptation aims to mitigate distribution shifts among different domains. However, traditional formulations are mostly limited to categorical domains, greatly simplifying nuanced domain relationships in the real world. In this work, we tackle a generalization with taxonomy-structured domains, which formalizes domains with nested, hierarchical similarity structures such as animal species and product catalogs. We build on the classic adversarial framework and introduce a novel taxonomist, which competes with the adversarial discriminator to preserve the taxonomy information. The equilibrium recovers the classic adversarial domain adaptation’s solution if given a non-informative domain taxonomy (e.g., a flat taxonomy where all leaf nodes connect to the root node) while yielding non-trivial results with other taxonomies. Empirically, our method achieves state-of-the-art performance on both synthetic and real-world datasets with successful adaptation.
APA
Liu, T., Xu, Z., He, H., Hao, G., Lee, G. & Wang, H.. (2023). Taxonomy-Structured Domain Adaptation. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:22215-22232 Available from https://proceedings.mlr.press/v202/liu23ap.html.

Related Material