{"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/towards-fairness-in-visual-recognition","title":"Towards Fairness in Visual Recognition: Effective Strategies for Bias Mitigation","arxiv_id":"1911.11834","date":"2019-11-26","proceeding":"CVPR 2020 6","authors":["Zeyu Wang","Klint Qinami","Ioannis Christos Karakozis","Kyle Genova","Prem Nair","Kenji Hata","Olga Russakovsky"],"abstract":"Computer vision models learn to perform a task by capturing relevant statistics from training data. It has been shown that models learn spurious age, gender, and race correlations when trained for seemingly unrelated tasks like activity recognition or image captioning. Various mitigation techniques have been presented to prevent models from utilizing or learning such biases. However, there has been little systematic comparison between these techniques. We design a simple but surprisingly effective visual recognition benchmark for studying bias mitigation. Using this benchmark, we provide a thorough analysis of a wide range of techniques. We highlight the shortcomings of popular adversarial training approaches for bias mitigation, propose a simple but similarly effective alternative to the inference-time Reducing Bias Amplification method of Zhao et al., and design a domain-independent training technique that outperforms all other methods. Finally, we validate our findings on the attribute classification task in the CelebA dataset, where attribute presence is known to be correlated with the gender of people in the image, and demonstrate that the proposed technique is effective at mitigating real-world gender bias.","url_abs":"https://arxiv.org/abs/1911.11834v2","url_pdf":"https://arxiv.org/pdf/1911.11834v2.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":"towards-fairness-in-visual-recognition","repo_url":"https://github.com/princetonvisualai/DomainBiasMitigation","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"towards-fairness-in-visual-recognition","repo_url":"https://github.com/junjie1003/DL-Fairness-Study","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"towards-fairness-in-visual-recognition","repo_url":"https://github.com/ys-zong/medfair","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"activity-recognition","task_name":"Activity Recognition"},{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"fairness","task_name":"Fairness"},{"task_slug":"image-captioning","task_name":"Image Captioning"},{"task_slug":"out-of-distribution-generalization","task_name":"Out-of-Distribution Generalization"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1911.11834","atlas_url":"https://app.syntology.ai/?focus=1911.11834","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1911.11834"}},"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/ys-zong/medfair","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/princetonvisualai/DomainBiasMitigation","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/junjie1003/DL-Fairness-Study","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":8,"unverified":4},"by_repo_kind":{"listed":{"samples":12,"ran":8,"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":"94cb2e164ab42106","entry":"accuracy_equality_difference","repo":"junjie1003/DL-Fairness-Study","repo_kind":"listed","path":"metrics.py","file_url":"https://github.com/junjie1003/DL-Fairness-Study/blob/HEAD/metrics.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"94cb2e164ab42106"}},{"code_sha256_prefix":"959c350e8c71f0f1","entry":"get_features","repo":"junjie1003/DL-Fairness-Study","repo_kind":"listed","path":"FLAC/get_imagenet_bias_features.py","file_url":"https://github.com/junjie1003/DL-Fairness-Study/blob/HEAD/FLAC/get_imagenet_bias_features.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"959c350e8c71f0f1"}},{"code_sha256_prefix":"7f91c1a83df06943","entry":"get_metric_list","repo":"junjie1003/DL-Fairness-Study","repo_kind":"listed","path":"helper.py","file_url":"https://github.com/junjie1003/DL-Fairness-Study/blob/HEAD/helper.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"7f91c1a83df06943"}},{"code_sha256_prefix":"cc8d2d0c85a28685","entry":"get_results","repo":"junjie1003/DL-Fairness-Study","repo_kind":"listed","path":"helper.py","file_url":"https://github.com/junjie1003/DL-Fairness-Study/blob/HEAD/helper.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"cc8d2d0c85a28685"}},{"code_sha256_prefix":"7504e58a0fedb5e9","entry":"prepare_cifar10s_data","repo":"junjie1003/DL-Fairness-Study","repo_kind":"listed","path":"FDR/utils.py","file_url":"https://github.com/junjie1003/DL-Fairness-Study/blob/HEAD/FDR/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"7504e58a0fedb5e9"}},{"code_sha256_prefix":"19dd1dafa5aab492","entry":"prepare_data","repo":"junjie1003/DL-Fairness-Study","repo_kind":"listed","path":"FDR/utils.py","file_url":"https://github.com/junjie1003/DL-Fairness-Study/blob/HEAD/FDR/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"19dd1dafa5aab492"}},{"code_sha256_prefix":"cb8230eee410ffa7","entry":"test","repo":"junjie1003/DL-Fairness-Study","repo_kind":"listed","path":"FAAP/train_deployed_model.py","file_url":"https://github.com/junjie1003/DL-Fairness-Study/blob/HEAD/FAAP/train_deployed_model.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"cb8230eee410ffa7"}},{"code_sha256_prefix":"d7fecd5203c15e82","entry":"train","repo":"junjie1003/DL-Fairness-Study","repo_kind":"listed","path":"FAAP/train_deployed_model.py","file_url":"https://github.com/junjie1003/DL-Fairness-Study/blob/HEAD/FAAP/train_deployed_model.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d7fecd5203c15e82"}},{"code_sha256_prefix":"828a99fb2a8fc736","entry":"get_cifar10s","repo":"junjie1003/DL-Fairness-Study","repo_kind":"listed","path":"FDR/cifar10s.py","file_url":"https://github.com/junjie1003/DL-Fairness-Study/blob/HEAD/FDR/cifar10s.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"828a99fb2a8fc736"}},{"code_sha256_prefix":"83822abaa0a90439","entry":"get_marginal","repo":"junjie1003/DL-Fairness-Study","repo_kind":"listed","path":"FLAC/get_imagenet_bias_features.py","file_url":"https://github.com/junjie1003/DL-Fairness-Study/blob/HEAD/FLAC/get_imagenet_bias_features.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"83822abaa0a90439"}},{"code_sha256_prefix":"23323ccad9b8e386","entry":"prepare_utkface_data","repo":"junjie1003/DL-Fairness-Study","repo_kind":"listed","path":"FDR/utils.py","file_url":"https://github.com/junjie1003/DL-Fairness-Study/blob/HEAD/FDR/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"23323ccad9b8e386"}},{"code_sha256_prefix":"f903d35c92c9ab78","entry":"valid_per_epoch","repo":"junjie1003/DL-Fairness-Study","repo_kind":"listed","path":"FDR/cifar10s.py","file_url":"https://github.com/junjie1003/DL-Fairness-Study/blob/HEAD/FDR/cifar10s.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f903d35c92c9ab78"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}