{"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/learning-to-learn-from-noisy-labeled-data","title":"Learning to Learn from Noisy Labeled Data","arxiv_id":"1812.05214","date":"2018-12-13","proceeding":"CVPR 2019 6","authors":["Junnan Li","Yongkang Wong","Qi Zhao","Mohan Kankanhalli"],"abstract":"Despite the success of deep neural networks (DNNs) in image classification\ntasks, the human-level performance relies on massive training data with\nhigh-quality manual annotations, which are expensive and time-consuming to\ncollect. There exist many inexpensive data sources on the web, but they tend to\ncontain inaccurate labels. Training on noisy labeled datasets causes\nperformance degradation because DNNs can easily overfit to the label noise. To\novercome this problem, we propose a noise-tolerant training algorithm, where a\nmeta-learning update is performed prior to conventional gradient update. The\nproposed meta-learning method simulates actual training by generating synthetic\nnoisy labels, and train the model such that after one gradient update using\neach set of synthetic noisy labels, the model does not overfit to the specific\nnoise. We conduct extensive experiments on the noisy CIFAR-10 dataset and the\nClothing1M dataset. The results demonstrate the advantageous performance of the\nproposed method compared to several state-of-the-art baselines.","url_abs":"http://arxiv.org/abs/1812.05214v2","url_pdf":"http://arxiv.org/pdf/1812.05214v2.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":"learning-to-learn-from-noisy-labeled-data","repo_url":"https://github.com/LiJunnan1992/MLNT","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"learning-with-noisy-labels","task_name":"Learning with noisy labels"},{"task_slug":"meta-learning","task_name":"Meta-Learning"},{"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":"MLNT","rank_in_archive_order":27,"of":51,"metrics":{"Accuracy":"73.47%"},"uses_additional_data":true}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1812.05214","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}