Optimal Learning from Label Proportions with General Loss Functions
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917233079353344 |
|---|---|
| author | Applebaum, Lorne Dick, Travis Gentile, Claudio Kaplan, Haim Koren, Tomer |
| author_facet | Applebaum, Lorne Dick, Travis Gentile, Claudio Kaplan, Haim Koren, Tomer |
| contents | Motivated by problems in online advertising, we address the task of Learning from Label Proportions (LLP). We introduce a novel and versatile low-variance debiasing methodology to learn from aggregate label information, significantly advancing the state of the art in LLP. Our debiasing approach exhibits remarkable flexibility, seamlessly accommodating a broad spectrum of practically relevant loss functions across both binary and multi-class classification settings. By carefully combining our estimators with standard techniques, we improve sample complexity guarantees for a large class of losses of practical relevance. We also empirically validate the efficacy of our proposed approach across a diverse array of benchmark datasets, demonstrating compelling empirical advantages over standard baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_15145 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Optimal Learning from Label Proportions with General Loss Functions Applebaum, Lorne Dick, Travis Gentile, Claudio Kaplan, Haim Koren, Tomer Machine Learning Motivated by problems in online advertising, we address the task of Learning from Label Proportions (LLP). We introduce a novel and versatile low-variance debiasing methodology to learn from aggregate label information, significantly advancing the state of the art in LLP. Our debiasing approach exhibits remarkable flexibility, seamlessly accommodating a broad spectrum of practically relevant loss functions across both binary and multi-class classification settings. By carefully combining our estimators with standard techniques, we improve sample complexity guarantees for a large class of losses of practical relevance. We also empirically validate the efficacy of our proposed approach across a diverse array of benchmark datasets, demonstrating compelling empirical advantages over standard baselines. |
| title | Optimal Learning from Label Proportions with General Loss Functions |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2509.15145 |