{"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/masking-a-new-perspective-of-noisy","title":"Masking: A New Perspective of Noisy Supervision","arxiv_id":"1805.08193","date":"2018-05-21","proceeding":"NeurIPS 2018 12","authors":["Bo Han","Jiangchao Yao","Gang Niu","Mingyuan Zhou","Ivor Tsang","Ya zhang","Masashi Sugiyama"],"abstract":"It is important to learn various types of classifiers given training data\nwith noisy labels. Noisy labels, in the most popular noise model hitherto, are\ncorrupted from ground-truth labels by an unknown noise transition matrix. Thus,\nby estimating this matrix, classifiers can escape from overfitting those noisy\nlabels. However, such estimation is practically difficult, due to either the\nindirect nature of two-step approaches, or not big enough data to afford\nend-to-end approaches. In this paper, we propose a human-assisted approach\ncalled Masking that conveys human cognition of invalid class transitions and\nnaturally speculates the structure of the noise transition matrix. To this end,\nwe derive a structure-aware probabilistic model incorporating a structure\nprior, and solve the challenges from structure extraction and structure\nalignment. Thanks to Masking, we only estimate unmasked noise transition\nprobabilities and the burden of estimation is tremendously reduced. We conduct\nextensive experiments on CIFAR-10 and CIFAR-100 with three noise structures as\nwell as the industrial-level Clothing1M with agnostic noise structure, and the\nresults show that Masking can improve the robustness of classifiers\nsignificantly.","url_abs":"http://arxiv.org/abs/1805.08193v2","url_pdf":"http://arxiv.org/pdf/1805.08193v2.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":"masking-a-new-perspective-of-noisy","repo_url":"https://github.com/bhanML/Masking","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"masking-a-new-perspective-of-noisy","repo_url":"https://github.com/bhanML/Co-teaching","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-classification-on-clothing1m","task":"Image Classification","dataset":"Clothing1M","model":"MASKING","rank_in_archive_order":43,"of":51,"metrics":{"Accuracy":"71.1%"},"uses_additional_data":true}],"syntology":{"syntology_url":"https://syntology.ai/paper/1805.08193","atlas_url":"https://app.syntology.ai/?focus=1805.08193","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1805.08193"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/bhanML/Co-teaching","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/bhanML/Masking","reach":null}],"summary":{"unverified":1},"by_repo_kind":{"official":{"samples":1,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"3a605c0419bb7124","entry":"dis_noise","repo":"bhanML/Masking","repo_kind":"official","path":"noise.py","file_url":"https://github.com/bhanML/Masking/blob/HEAD/noise.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3a605c0419bb7124"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}