# Learning from Label Proportions: A Mutual Contamination Framework

arXiv: https://arxiv.org/abs/2006.07330 · HTML: https://mlsanity.com/paper/2006.07330/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Jun 2020.

## Built on by (most cited first, 6 of 6)

- [Binary Classification from Multiple Unlabeled Datasets via Surrogate Set Classification](https://mlsanity.com/paper/2102.00678/) (2021)
- [Learning crop type mapping from regional label proportions in large-scale SAR and optical imagery](https://mlsanity.com/paper/2208.11607/) (2022)
- [Learning under Label Proportions for Text Classification](https://mlsanity.com/paper/2310.11707/) (2023)
- [Enhancing Self-Supervised Learning for Remote Sensing with Elevation Data: A Case Study with Scarce And High Level Semantic Labels](https://mlsanity.com/paper/2304.06857/) (2023)
- [AUC Optimization from Multiple Unlabeled Datasets](https://mlsanity.com/paper/2305.15776/) (2023)
- [Evaluating LLP Methods: Challenges and Approaches](https://mlsanity.com/paper/2310.19065/) (2023)

## Abstract (arXiv)

Learning from label proportions (LLP) is a weakly supervised setting for classification in which unlabeled training instances are grouped into bags, and each bag is annotated with the proportion of each class occurring in that bag. Prior work on LLP has yet to establish a consistent learning procedure, nor does there exist a theoretically justified, general purpose training criterion. In this work we address these two issues by posing LLP in terms of mutual contamination models (MCMs), which have recently been applied successfully to study various other weak supervision settings. In the process, we establish several novel technical results for MCMs, including unbiased losses and generalization error bounds under non-iid sampling plans. We also point out the limitations of a common experimental setting for LLP, and propose a new one based on our MCM framework.