Diversity-enhancing Generative Network for Few-shot Hypothesis Adaptation

Ruijiang Dong, Feng Liu, Haoang Chi, Tongliang Liu, Mingming Gong, Gang Niu, Masashi Sugiyama, Bo Han
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:8260-8275, 2023.

Abstract

Generating unlabeled data has been recently shown to help address the few-shot hypothesis adaptation (FHA) problem, where we aim to train a classifier for the target domain with a few labeled target-domain data and a well-trained source-domain classifier (i.e., a source hypothesis), for the additional information of the highly-compatible unlabeled data. However, the generated data of the existing methods are extremely similar or even the same. The strong dependency among the generated data will lead the learning to fail. In this paper, we propose a diversity-enhancing generative network (DEG-Net) for the FHA problem, which can generate diverse unlabeled data with the help of a kernel independence measure: the Hilbert-Schmidt independence criterion (HSIC). Specifically, DEG-Net will generate data via minimizing the HSIC value (i.e., maximizing the independence) among the semantic features of the generated data. By DEG-Net, the generated unlabeled data are more diverse and more effective for addressing the FHA problem. Experimental results show that the DEG-Net outperforms existing FHA baselines and further verifies that generating diverse data plays an important role in addressing the FHA problem.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-dong23d, title = {Diversity-enhancing Generative Network for Few-shot Hypothesis Adaptation}, author = {Dong, Ruijiang and Liu, Feng and Chi, Haoang and Liu, Tongliang and Gong, Mingming and Niu, Gang and Sugiyama, Masashi and Han, Bo}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {8260--8275}, 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/dong23d/dong23d.pdf}, url = {https://proceedings.mlr.press/v202/dong23d.html}, abstract = {Generating unlabeled data has been recently shown to help address the few-shot hypothesis adaptation (FHA) problem, where we aim to train a classifier for the target domain with a few labeled target-domain data and a well-trained source-domain classifier (i.e., a source hypothesis), for the additional information of the highly-compatible unlabeled data. However, the generated data of the existing methods are extremely similar or even the same. The strong dependency among the generated data will lead the learning to fail. In this paper, we propose a diversity-enhancing generative network (DEG-Net) for the FHA problem, which can generate diverse unlabeled data with the help of a kernel independence measure: the Hilbert-Schmidt independence criterion (HSIC). Specifically, DEG-Net will generate data via minimizing the HSIC value (i.e., maximizing the independence) among the semantic features of the generated data. By DEG-Net, the generated unlabeled data are more diverse and more effective for addressing the FHA problem. Experimental results show that the DEG-Net outperforms existing FHA baselines and further verifies that generating diverse data plays an important role in addressing the FHA problem.} }
Endnote
%0 Conference Paper %T Diversity-enhancing Generative Network for Few-shot Hypothesis Adaptation %A Ruijiang Dong %A Feng Liu %A Haoang Chi %A Tongliang Liu %A Mingming Gong %A Gang Niu %A Masashi Sugiyama %A Bo Han %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-dong23d %I PMLR %P 8260--8275 %U https://proceedings.mlr.press/v202/dong23d.html %V 202 %X Generating unlabeled data has been recently shown to help address the few-shot hypothesis adaptation (FHA) problem, where we aim to train a classifier for the target domain with a few labeled target-domain data and a well-trained source-domain classifier (i.e., a source hypothesis), for the additional information of the highly-compatible unlabeled data. However, the generated data of the existing methods are extremely similar or even the same. The strong dependency among the generated data will lead the learning to fail. In this paper, we propose a diversity-enhancing generative network (DEG-Net) for the FHA problem, which can generate diverse unlabeled data with the help of a kernel independence measure: the Hilbert-Schmidt independence criterion (HSIC). Specifically, DEG-Net will generate data via minimizing the HSIC value (i.e., maximizing the independence) among the semantic features of the generated data. By DEG-Net, the generated unlabeled data are more diverse and more effective for addressing the FHA problem. Experimental results show that the DEG-Net outperforms existing FHA baselines and further verifies that generating diverse data plays an important role in addressing the FHA problem.
APA
Dong, R., Liu, F., Chi, H., Liu, T., Gong, M., Niu, G., Sugiyama, M. & Han, B.. (2023). Diversity-enhancing Generative Network for Few-shot Hypothesis Adaptation. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:8260-8275 Available from https://proceedings.mlr.press/v202/dong23d.html.

Related Material