Unmixing Mean Embeddings for Domain Adaptation with Target Label Proportion

Alain Rakotomamonjy, Maxime Berar, Mokhtar Z. Alaya
Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, PMLR 300:3214-3222, 2026.

Abstract

We introduce a novel approach to domain adaptation within the context of Learning from Label Proportions (LLP). We address the challenging scenario where labeled samples are available in the source domain, but only bags of unlabeled samples with their corresponding label proportions are accessible in the target domain. Our proposed method, bagMME (Bag Matching Mean Embeddings), tackles the distributional shift between domains by focusing on matching class-conditional distributions. A key contribution of bagMME is a simple yet effective unmixing strategy that leverages the target label proportions to estimate the target class-conditional mean embeddings. These estimated target means are then aligned with their corresponding source class-conditional means, thereby reducing the domain discrepancy. We theoretically demonstrate the soundness of our approach and its effectiveness in mitigating distributional shifts. Extensive experiments on various computer vision datasets showcase the superior performance of bagMME compared to state-of-the-art baselines. Our results highlight the critical role of incorporating target label proportions into the learning process for improved generalization on the target domain.

Cite this Paper

BibTeX
@InProceedings{pmlr-v300-rakotomamonjy26a, title = { Unmixing Mean Embeddings for Domain Adaptation with Target Label Proportion }, author = {Rakotomamonjy, Alain and Berar, Maxime and Alaya, Mokhtar Z.}, booktitle = {Proceedings of The 29th International Conference on Artificial Intelligence and Statistics}, pages = {3214--3222}, year = {2026}, editor = {Khan, Emtiyaz and Li, Yingzhen and Solin, Arno and Ramdas, Aaditya}, volume = {300}, series = {Proceedings of Machine Learning Research}, month = {02--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v300/main/assets/rakotomamonjy26a/rakotomamonjy26a.pdf}, url = {https://proceedings.mlr.press/v300/rakotomamonjy26a.html}, abstract = { We introduce a novel approach to domain adaptation within the context of Learning from Label Proportions (LLP). We address the challenging scenario where labeled samples are available in the source domain, but only bags of unlabeled samples with their corresponding label proportions are accessible in the target domain. Our proposed method, bagMME (Bag Matching Mean Embeddings), tackles the distributional shift between domains by focusing on matching class-conditional distributions. A key contribution of bagMME is a simple yet effective unmixing strategy that leverages the target label proportions to estimate the target class-conditional mean embeddings. These estimated target means are then aligned with their corresponding source class-conditional means, thereby reducing the domain discrepancy. We theoretically demonstrate the soundness of our approach and its effectiveness in mitigating distributional shifts. Extensive experiments on various computer vision datasets showcase the superior performance of bagMME compared to state-of-the-art baselines. Our results highlight the critical role of incorporating target label proportions into the learning process for improved generalization on the target domain. } }
Endnote
%0 Conference Paper %T Unmixing Mean Embeddings for Domain Adaptation with Target Label Proportion %A Alain Rakotomamonjy %A Maxime Berar %A Mokhtar Z. Alaya %B Proceedings of The 29th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2026 %E Emtiyaz Khan %E Yingzhen Li %E Arno Solin %E Aaditya Ramdas %F pmlr-v300-rakotomamonjy26a %I PMLR %P 3214--3222 %U https://proceedings.mlr.press/v300/rakotomamonjy26a.html %V 300 %X We introduce a novel approach to domain adaptation within the context of Learning from Label Proportions (LLP). We address the challenging scenario where labeled samples are available in the source domain, but only bags of unlabeled samples with their corresponding label proportions are accessible in the target domain. Our proposed method, bagMME (Bag Matching Mean Embeddings), tackles the distributional shift between domains by focusing on matching class-conditional distributions. A key contribution of bagMME is a simple yet effective unmixing strategy that leverages the target label proportions to estimate the target class-conditional mean embeddings. These estimated target means are then aligned with their corresponding source class-conditional means, thereby reducing the domain discrepancy. We theoretically demonstrate the soundness of our approach and its effectiveness in mitigating distributional shifts. Extensive experiments on various computer vision datasets showcase the superior performance of bagMME compared to state-of-the-art baselines. Our results highlight the critical role of incorporating target label proportions into the learning process for improved generalization on the target domain.
APA
Rakotomamonjy, A., Berar, M. & Alaya, M.Z.. (2026). Unmixing Mean Embeddings for Domain Adaptation with Target Label Proportion . Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 300:3214-3222 Available from https://proceedings.mlr.press/v300/rakotomamonjy26a.html.

Related Material