{"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/preventing-manifold-intrusion-with-locality","title":"Preventing Manifold Intrusion with Locality: Local Mixup","arxiv_id":"2201.04368","date":"2022-01-12","proceeding":null,"authors":["Raphael Baena","Lucas Drumetz","Vincent Gripon"],"abstract":"Mixup is a data-dependent regularization technique that consists in linearly interpolating input samples and associated outputs. It has been shown to improve accuracy when used to train on standard machine learning datasets. However, authors have pointed out that Mixup can produce out-of-distribution virtual samples and even contradictions in the augmented training set, potentially resulting in adversarial effects. In this paper, we introduce Local Mixup in which distant input samples are weighted down when computing the loss. In constrained settings we demonstrate that Local Mixup can create a trade-off between bias and variance, with the extreme cases reducing to vanilla training and classical Mixup. Using standardized computer vision benchmarks , we also show that Local Mixup can improve test accuracy.","url_abs":"https://arxiv.org/abs/2201.04368v1","url_pdf":"https://arxiv.org/pdf/2201.04368v1.pdf","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":"preventing-manifold-intrusion-with-locality","repo_url":"https://github.com/raphael-baena/Local-Mixup","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-2.0"}}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"}],"methods":[{"method_slug":"local-mixup","method_name":"Local Mixup"},{"method_slug":"mixup","method_name":"Mixup"}],"datasets_introduced":[{"slug":"two-coiling-spiral","name":"Two Coiling Spirals","full_name":""}],"methods_introduced":[{"slug":"local-mixup","name":"Local Mixup","full_name":"Local Mixup"}],"results":[{"leaderboard":"/sota/image-classification-on-cifar-10","task":"Image Classification","dataset":"CIFAR-10","model":"Local Mixup Resnet18","rank_in_archive_order":124,"of":265,"metrics":{"Percentage correct":"95.97"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-fashion-mnist","task":"Image Classification","dataset":"Fashion-MNIST","model":"Local Mixup DenseNet","rank_in_archive_order":7,"of":34,"metrics":{"Percentage error":"5.97"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-svhn","task":"Image Classification","dataset":"SVHN","model":"Local Mixup LeNet","rank_in_archive_order":45,"of":62,"metrics":{"Percentage error":"8.20"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2201.04368","atlas_url":"https://app.syntology.ai/?focus=2201.04368","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}