{"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/maximum-classifier-discrepancy-for","title":"Maximum Classifier Discrepancy for Unsupervised Domain Adaptation","arxiv_id":"1712.02560","date":"2017-12-07","proceeding":"CVPR 2018 6","authors":["Kuniaki Saito","Kohei Watanabe","Yoshitaka Ushiku","Tatsuya Harada"],"abstract":"In this work, we present a method for unsupervised domain adaptation. Many\nadversarial learning methods train domain classifier networks to distinguish\nthe features as either a source or target and train a feature generator network\nto mimic the discriminator. Two problems exist with these methods. First, the\ndomain classifier only tries to distinguish the features as a source or target\nand thus does not consider task-specific decision boundaries between classes.\nTherefore, a trained generator can generate ambiguous features near class\nboundaries. Second, these methods aim to completely match the feature\ndistributions between different domains, which is difficult because of each\ndomain's characteristics.\n  To solve these problems, we introduce a new approach that attempts to align\ndistributions of source and target by utilizing the task-specific decision\nboundaries. We propose to maximize the discrepancy between two classifiers'\noutputs to detect target samples that are far from the support of the source. A\nfeature generator learns to generate target features near the support to\nminimize the discrepancy. Our method outperforms other methods on several\ndatasets of image classification and semantic segmentation. The codes are\navailable at \\url{https://github.com/mil-tokyo/MCD_DA}","url_abs":"http://arxiv.org/abs/1712.02560v4","url_pdf":"http://arxiv.org/pdf/1712.02560v4.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":"maximum-classifier-discrepancy-for","repo_url":"https://github.com/mil-tokyo/MCD_DA","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"maximum-classifier-discrepancy-for","repo_url":"https://github.com/adapt-python/adapt","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"maximum-classifier-discrepancy-for","repo_url":"https://github.com/mcd4874/neurips_competition","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"maximum-classifier-discrepancy-for","repo_url":"https://github.com/onedayatatime0923/Cycle_Mcd_Gan","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"maximum-classifier-discrepancy-for","repo_url":"https://github.com/thuml/Transfer-Learning-Library","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"maximum-classifier-discrepancy-for","repo_url":"https://github.com/Elman295/Maximum-Classifier-Discrepancy-for-Unsupervised-Domain-Adaptation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"maximum-classifier-discrepancy-for","repo_url":"https://github.com/HiGal/Domain-Adaptation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"maximum-classifier-discrepancy-for","repo_url":"https://github.com/Nyn-ynu/MCD","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"maximum-classifier-discrepancy-for","repo_url":"https://github.com/kevinmusgrave/pytorch-adapt","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"multi-source-unsupervised-domain-adaptation","task_name":"Multi-Source Unsupervised Domain Adaptation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"unsupervised-domain-adaptation","task_name":"Unsupervised Domain Adaptation"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/domain-adaptation-on-hmdbfull-to-ucf","task":"Domain Adaptation","dataset":"HMDBfull-to-UCF","model":"MCD","rank_in_archive_order":4,"of":5,"metrics":{"Accuracy":"79.34"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-mnist-to-usps","task":"Domain Adaptation","dataset":"MNIST-to-USPS","model":"MCD","rank_in_archive_order":13,"of":14,"metrics":{"Accuracy":"93.8"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-svhn-to-mnist","task":"Domain Adaptation","dataset":"SVHN-to-MNIST","model":"MCD","rank_in_archive_order":7,"of":14,"metrics":{"Accuracy":"95.8"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-synsig-to-gtsrb","task":"Domain Adaptation","dataset":"SYNSIG-to-GTSRB","model":"MCD","rank_in_archive_order":3,"of":6,"metrics":{"Accuracy":"94.4"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-ucf-to-hmdbfull","task":"Domain Adaptation","dataset":"UCF-to-HMDBfull","model":"MCD","rank_in_archive_order":5,"of":5,"metrics":{"Accuracy":"73.89"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-usps-to-mnist","task":"Domain Adaptation","dataset":"USPS-to-MNIST","model":"MCD","rank_in_archive_order":13,"of":14,"metrics":{"Accuracy":"95.7"},"uses_additional_data":false},{"leaderboard":"/sota/synthetic-to-real-translation-on-syn2real-c","task":"Synthetic-to-Real Translation","dataset":"Syn2Real-C","model":"MCD","rank_in_archive_order":4,"of":6,"metrics":{"Accuracy":"71.9"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1712.02560","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1712.02560"}},"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/Nyn-ynu/MCD","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/adapt-python/adapt","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mcd4874/neurips_competition","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/kevinmusgrave/pytorch-adapt","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/thuml/Transfer-Learning-Library","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Elman295/Maximum-Classifier-Discrepancy-for-Unsupervised-Domain-Adaptation","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/HiGal/Domain-Adaptation","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mil-tokyo/MCD_DA","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/onedayatatime0923/Cycle_Mcd_Gan","reach":{"status":"unanswered"}}],"summary":{"ran_honours":1},"by_repo_kind":{"listed":{"samples":1,"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":1,"samples":[{"code_sha256_prefix":"92b5f7c292e95c2c","entry":"test","repo":"HiGal/Domain-Adaptation","repo_kind":"listed","path":"baseline.py","file_url":"https://github.com/HiGal/Domain-Adaptation/blob/HEAD/baseline.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"92b5f7c292e95c2c"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}