{"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/normalized-loss-functions-for-deep-learning","title":"Normalized Loss Functions for Deep Learning with Noisy Labels","arxiv_id":"2006.13554","date":"2020-06-24","proceeding":"ICML 2020 1","authors":["Xingjun Ma","Hanxun Huang","Yisen Wang","Simone Romano","Sarah Erfani","James Bailey"],"abstract":"Robust loss functions are essential for training accurate deep neural networks (DNNs) in the presence of noisy (incorrect) labels. It has been shown that the commonly used Cross Entropy (CE) loss is not robust to noisy labels. Whilst new loss functions have been designed, they are only partially robust. In this paper, we theoretically show by applying a simple normalization that: any loss can be made robust to noisy labels. However, in practice, simply being robust is not sufficient for a loss function to train accurate DNNs. By investigating several robust loss functions, we find that they suffer from a problem of underfitting. To address this, we propose a framework to build robust loss functions called Active Passive Loss (APL). APL combines two robust loss functions that mutually boost each other. Experiments on benchmark datasets demonstrate that the family of new loss functions created by our APL framework can consistently outperform state-of-the-art methods by large margins, especially under large noise rates such as 60% or 80% incorrect labels.","url_abs":"https://arxiv.org/abs/2006.13554v1","url_pdf":"https://arxiv.org/pdf/2006.13554v1.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":"normalized-loss-functions-for-deep-learning","repo_url":"https://github.com/HanxunH/Active-Passive-Losses","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"normalized-loss-functions-for-deep-learning","repo_url":"https://github.com/XinshaoAmosWang/DerivativeManipulation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"normalized-loss-functions-for-deep-learning","repo_url":"https://github.com/XinshaoAmosWang/Emphasis-Regularisation-by-Gradient-Rescaling","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"normalized-loss-functions-for-deep-learning","repo_url":"https://github.com/XinshaoAmosWang/Improving-Mean-Absolute-Error-against-CCE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"deep-learning","task_name":"Deep Learning"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"learning-with-noisy-labels","task_name":"Learning with noisy labels"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-classification-on-mini-webvision-1-0","task":"Image Classification","dataset":"mini WebVision 1.0","model":"NCE+RCE (ResNet-50)","rank_in_archive_order":45,"of":47,"metrics":{"ImageNet Top-1 Accuracy":"62.64"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-mini-webvision-1-0","task":"Image Classification","dataset":"mini WebVision 1.0","model":"NCE+MAE (ResNet-50)","rank_in_archive_order":46,"of":47,"metrics":{"ImageNet Top-1 Accuracy":"62.36"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2006.13554","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2006.13554"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+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/XinshaoAmosWang/Emphasis-Regularisation-by-Gradient-Rescaling","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/XinshaoAmosWang/DerivativeManipulation","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/XinshaoAmosWang/Improving-Mean-Absolute-Error-against-CCE","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/HanxunH/Active-Passive-Losses","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":1,"unverified":11},"by_repo_kind":{"official":{"samples":12,"ran":1,"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":0,"samples":[{"code_sha256_prefix":"3a4f8ebc3e6cbd58","entry":"setup_logger","repo":"HanxunH/Active-Passive-Losses","repo_kind":"official","path":"util.py","file_url":"https://github.com/HanxunH/Active-Passive-Losses/blob/HEAD/util.py","link_basis":"plan_row","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3a4f8ebc3e6cbd58"}},{"code_sha256_prefix":"268340d6b6a411c7","entry":"ResNet18","repo":"HanxunH/Active-Passive-Losses","repo_kind":"official","path":"archive/model.py","file_url":"https://github.com/HanxunH/Active-Passive-Losses/blob/HEAD/archive/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"268340d6b6a411c7"}},{"code_sha256_prefix":"18a4a9f249bb889a","entry":"ResNet34","repo":"HanxunH/Active-Passive-Losses","repo_kind":"official","path":"archive/model.py","file_url":"https://github.com/HanxunH/Active-Passive-Losses/blob/HEAD/archive/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"18a4a9f249bb889a"}},{"code_sha256_prefix":"30165ab348cc2c93","entry":"ResNet50","repo":"HanxunH/Active-Passive-Losses","repo_kind":"official","path":"archive/model.py","file_url":"https://github.com/HanxunH/Active-Passive-Losses/blob/HEAD/archive/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"30165ab348cc2c93"}},{"code_sha256_prefix":"ed42eca6cf2baf78","entry":"accuracy","repo":"HanxunH/Active-Passive-Losses","repo_kind":"official","path":"util.py","file_url":"https://github.com/HanxunH/Active-Passive-Losses/blob/HEAD/util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ed42eca6cf2baf78"}},{"code_sha256_prefix":"53636482246a0fcc","entry":"accuracy","repo":"HanxunH/Active-Passive-Losses","repo_kind":"official","path":"archive/utils/utils.py","file_url":"https://github.com/HanxunH/Active-Passive-Losses/blob/HEAD/archive/utils/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"53636482246a0fcc"}},{"code_sha256_prefix":"baf477efd17d8a86","entry":"build_for_cifar100","repo":"HanxunH/Active-Passive-Losses","repo_kind":"official","path":"dataset.py","file_url":"https://github.com/HanxunH/Active-Passive-Losses/blob/HEAD/dataset.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"baf477efd17d8a86"}},{"code_sha256_prefix":"a42ae792b2745b93","entry":"count_parameters_in_MB","repo":"HanxunH/Active-Passive-Losses","repo_kind":"official","path":"archive/utils/utils.py","file_url":"https://github.com/HanxunH/Active-Passive-Losses/blob/HEAD/archive/utils/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a42ae792b2745b93"}},{"code_sha256_prefix":"eaac814ea08816f7","entry":"load_checkpoint_history","repo":"HanxunH/Active-Passive-Losses","repo_kind":"official","path":"collect_exp_results.py","file_url":"https://github.com/HanxunH/Active-Passive-Losses/blob/HEAD/collect_exp_results.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"eaac814ea08816f7"}},{"code_sha256_prefix":"5ef20ee93730c46b","entry":"log_display","repo":"HanxunH/Active-Passive-Losses","repo_kind":"official","path":"util.py","file_url":"https://github.com/HanxunH/Active-Passive-Losses/blob/HEAD/util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5ef20ee93730c46b"}},{"code_sha256_prefix":"b175d19a7622c5c6","entry":"multiclass_noisify","repo":"HanxunH/Active-Passive-Losses","repo_kind":"official","path":"dataset.py","file_url":"https://github.com/HanxunH/Active-Passive-Losses/blob/HEAD/dataset.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b175d19a7622c5c6"}},{"code_sha256_prefix":"542dbec466e02973","entry":"other_class","repo":"HanxunH/Active-Passive-Losses","repo_kind":"official","path":"dataset.py","file_url":"https://github.com/HanxunH/Active-Passive-Losses/blob/HEAD/dataset.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"542dbec466e02973"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}