{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/pue-biased-positive-unlabeled-learning","title":"PUe: Biased Positive-Unlabeled Learning Enhancement by Causal Inference","arxiv_id":null,"date":"2023-09-21","proceeding":"NeurIPS 2023 11","authors":[],"abstract":"Positive-Unlabeled (PU) learning aims to achieve high-accuracy binary classification with \nlimited labeled positive examples and numerous unlabeled ones. Existing cost-sensitive-based \nmethods often rely on strong assumptions that examples with an observed positive label were \nselected entirely at random. In fact, the uneven distribution of labels is prevalent in \nreal-world PU problems, indicating that most actual positive and unlabeled data are subject \nto selection bias. In this paper, we propose a PU learning enhancement (PUe) algorithm \nbased on causal inference theory, which employs normalized propensity scores and normalized \ninverse probability weighting (NIPW) techniques to reconstruct the loss function, thus \nobtaining a consistent, unbiased estimate of the classifier and enhancing the model's \nperformance. Moreover, we investigate and propose a method for estimating propensity scores \nin deep learning using regularization techniques when the labeling mechanism is unknown. \nOur experiments on three benchmark datasets demonstrate the proposed PUe algorithm significantly \nimproves the accuracy of classifiers on non-uniform label distribution datasets compared to \nadvanced cost-sensitive PU methods. Codes are available at https://github.com/huawei-noah/Noah-research/tree/master/PUe and https://gitee.com/mindspore/models/tree/master/research/cv/PUe.","url_abs":"https://openreview.net/forum?id=6vtZIoxZoJ","url_pdf":"https://openreview.net/pdf?id=6vtZIoxZoJ","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"pue-biased-positive-unlabeled-learning","repo_url":"https://github.com/huawei-noah/noah-research","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[],"methods":[{"method_slug":"causal-inference","method_name":"Causal inference"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}