{"type":"ScholarlyArticle","title":"Two-stage Training for Learning from Label Proportions","authors":[{"name":"Jiabin Liu","slug":"jiabin-liu","given":"Jiabin","family":"Liu","affiliation":""},{"name":"Bo Wang","slug":"bo-wang","given":"Bo","family":"Wang","affiliation":""},{"name":"Xin Shen","slug":"xin-shen","given":"Xin","family":"Shen","affiliation":""},{"name":"Zhiquan Qi","slug":"zhiquan-qi","given":"Zhiquan","family":"Qi","affiliation":""},{"name":"Yingjie Tian","slug":"yingjie-tian","given":"Yingjie","family":"Tian","affiliation":""}],"abstract":"Learning from label proportions (LLP) aims at learning an instance-level classifier with label proportions in grouped training data. Existing deep learning based LLP methods utilize end-to-end pipelines to obtain the proportional loss with Kullback-Leibler divergence between the bag-level prior and posterior class distributions. However, the unconstrained optimization on this objective can hardly reach a solution in accordance with the given proportions. Besides, concerning the probabilistic classifier, this strategy unavoidably results in high-entropy conditional class distributions at the instance level. These issues further degrade the performance of the instance-level classification. In this paper, we regard these problems as noisy pseudo labeling, and instead impose the strict proportion consistency on the classifier with a constrained optimization as a continuous training stage for existing LLP classifiers. In addition, we introduce the mixup strategy and symmetric cross-entropy to further reduce the label noise. Our framework is model-agnostic, and demonstrates compelling performance improvement in extensive experiments, when incorporated into other deep LLP models as a post-hoc phase.","doi":"10.24963/ijcai.2021/377","url":"https://ijcai.org/proceedings/2021/377","pdf":"https://ijcai.org/proceedings/2021/0377.pdf","markdown":"https://ijcai.org/proceedings/2021/377.md","bibtex":"https://ijcai.org/proceedings/2021/bibtex/377","volume":{"slug":"2021","title":"Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2021","url":"https://ijcai.org/proceedings/2021/","year":2021,"location":"Montreal","dates":"19-27 August 2021","editors":"Prof. Zhi-Hua Zhou","isbn":"978-0-9992411-9-6","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2021"},"section":"Machine Learning","track":"Main Track","number":377,"pages":{"first":2737,"last":2743},"keywords":["Machine Learning: Weakly Supervised Learning","Machine Learning: Deep Learning","Machine Learning: Classification"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}