{"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/proxynca-revisiting-and-revitalizing-proxy","title":"ProxyNCA++: Revisiting and Revitalizing Proxy Neighborhood Component Analysis","arxiv_id":"2004.01113","date":"2020-04-02","proceeding":"ECCV 2020 8","authors":["Eu Wern Teh","Terrance DeVries","Graham W. Taylor"],"abstract":"We consider the problem of distance metric learning (DML), where the task is to learn an effective similarity measure between images. We revisit ProxyNCA and incorporate several enhancements. We find that low temperature scaling is a performance-critical component and explain why it works. Besides, we also discover that Global Max Pooling works better in general when compared to Global Average Pooling. Additionally, our proposed fast moving proxies also addresses small gradient issue of proxies, and this component synergizes well with low temperature scaling and Global Max Pooling. Our enhanced model, called ProxyNCA++, achieves a 22.9 percentage point average improvement of Recall@1 across four different zero-shot retrieval datasets compared to the original ProxyNCA algorithm. Furthermore, we achieve state-of-the-art results on the CUB200, Cars196, Sop, and InShop datasets, achieving Recall@1 scores of 72.2, 90.1, 81.4, and 90.9, respectively.","url_abs":"https://arxiv.org/abs/2004.01113v2","url_pdf":"https://arxiv.org/pdf/2004.01113v2.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":"proxynca-revisiting-and-revitalizing-proxy","repo_url":"https://github.com/euwern/proxynca_pp","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"image-retrieval","task_name":"Image Retrieval"},{"task_slug":"metric-learning","task_name":"Metric Learning"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"max-pooling","method_name":"Max Pooling"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-retrieval-on-cars196","task":"Image Retrieval","dataset":"CARS196","model":"ProxyNCA++","rank_in_archive_order":2,"of":8,"metrics":{"R@1":"90.1"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-cub-200-2011","task":"Image Retrieval","dataset":"CUB-200-2011","model":"ProxyNCA++","rank_in_archive_order":3,"of":8,"metrics":{"R@1":"72.2"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-in-shop","task":"Image Retrieval","dataset":"In-Shop","model":"ProxyNCA++","rank_in_archive_order":3,"of":7,"metrics":{"R@1":"90.9"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-sop","task":"Image Retrieval","dataset":"SOP","model":"ProxyNCA++","rank_in_archive_order":5,"of":14,"metrics":{"R@1":"81.4"},"uses_additional_data":false},{"leaderboard":"/sota/metric-learning-on-cars196","task":"Metric Learning","dataset":"CARS196","model":"ResNet-50 + ProxyNCA++","rank_in_archive_order":23,"of":36,"metrics":{"R@1":"86.5"},"uses_additional_data":true},{"leaderboard":"/sota/metric-learning-on-cub-200-2011","task":"Metric Learning","dataset":"CUB-200-2011","model":"ResNet-50 + ProxyNCA++","rank_in_archive_order":15,"of":30,"metrics":{"R@1":"69.0"},"uses_additional_data":true},{"leaderboard":"/sota/metric-learning-on-in-shop-1","task":"Metric Learning","dataset":"In-Shop","model":"ResNet-50 + ProxyNCA++","rank_in_archive_order":12,"of":15,"metrics":{"R@1":"90.9"},"uses_additional_data":true},{"leaderboard":"/sota/metric-learning-on-stanford-online-products-1","task":"Metric Learning","dataset":"Stanford Online Products","model":"ResNet-50 + ProxyNCA++","rank_in_archive_order":21,"of":33,"metrics":{"R@1":"80.7"},"uses_additional_data":true}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2004.01113","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2004.01113"}},"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/euwern/proxynca_pp","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":5},"by_repo_kind":{"listed":{"samples":5,"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":0,"samples":[{"code_sha256_prefix":"cff1ce4645740206","entry":"binarize_and_smooth_labels","repo":"euwern/proxynca_pp","repo_kind":"listed","path":"loss.py","file_url":"https://github.com/euwern/proxynca_pp/blob/HEAD/loss.py","link_basis":"harvester_set","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":"cff1ce4645740206"}},{"code_sha256_prefix":"0d1bc17f353a5a5e","entry":"load_config","repo":"euwern/proxynca_pp","repo_kind":"listed","path":"utils.py","file_url":"https://github.com/euwern/proxynca_pp/blob/HEAD/utils.py","link_basis":"harvester_set","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":"0d1bc17f353a5a5e"}},{"code_sha256_prefix":"ecd1d401911526d0","entry":"pairwise_distance","repo":"euwern/proxynca_pp","repo_kind":"listed","path":"similarity.py","file_url":"https://github.com/euwern/proxynca_pp/blob/HEAD/similarity.py","link_basis":"harvester_set","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":"ecd1d401911526d0"}},{"code_sha256_prefix":"c89e3e64eac56c8b","entry":"predict_batchwise","repo":"euwern/proxynca_pp","repo_kind":"listed","path":"utils.py","file_url":"https://github.com/euwern/proxynca_pp/blob/HEAD/utils.py","link_basis":"harvester_set","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":"c89e3e64eac56c8b"}},{"code_sha256_prefix":"6d8f0f85042ae81c","entry":"predict_batchwise_inshop","repo":"euwern/proxynca_pp","repo_kind":"listed","path":"utils.py","file_url":"https://github.com/euwern/proxynca_pp/blob/HEAD/utils.py","link_basis":"harvester_set","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":"6d8f0f85042ae81c"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}