{"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/active-learning-for-convolutional-neural","title":"Active Learning for Convolutional Neural Networks: A Core-Set Approach","arxiv_id":"1708.00489","date":"2017-08-01","proceeding":"ICLR 2018 1","authors":["Ozan Sener","Silvio Savarese"],"abstract":"Convolutional neural networks (CNNs) have been successfully applied to many\nrecognition and learning tasks using a universal recipe; training a deep model\non a very large dataset of supervised examples. However, this approach is\nrather restrictive in practice since collecting a large set of labeled images\nis very expensive. One way to ease this problem is coming up with smart ways\nfor choosing images to be labelled from a very large collection (ie. active\nlearning).\n  Our empirical study suggests that many of the active learning heuristics in\nthe literature are not effective when applied to CNNs in batch setting.\nInspired by these limitations, we define the problem of active learning as\ncore-set selection, ie. choosing set of points such that a model learned over\nthe selected subset is competitive for the remaining data points. We further\npresent a theoretical result characterizing the performance of any selected\nsubset using the geometry of the datapoints. As an active learning algorithm,\nwe choose the subset which is expected to yield best result according to our\ncharacterization. Our experiments show that the proposed method significantly\noutperforms existing approaches in image classification experiments by a large\nmargin.","url_abs":"http://arxiv.org/abs/1708.00489v4","url_pdf":"http://arxiv.org/pdf/1708.00489v4.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":"active-learning-for-convolutional-neural","repo_url":"https://github.com/ozansener/active_learning_coreset","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"active-learning-for-convolutional-neural","repo_url":"https://github.com/blackhc/active-bayesian-coresets","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"active-learning-for-convolutional-neural","repo_url":"https://github.com/dsba-lab/openal","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"active-learning-for-convolutional-neural","repo_url":"https://github.com/hillup/active_learning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"active-learning-for-convolutional-neural","repo_url":"https://github.com/humanlab/rare-class-AL","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"active-learning-for-convolutional-neural","repo_url":"https://github.com/meghshukla/activelearningforhumanpose","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"AGPL-3.0"}},{"paper_slug":"active-learning-for-convolutional-neural","repo_url":"https://github.com/meghshukla/math-analysis-learningloss","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"AGPL-3.0"}},{"paper_slug":"active-learning-for-convolutional-neural","repo_url":"https://github.com/razvancaramalau/Sequential-GCN-for-Active-Learning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"active-learning-for-convolutional-neural","repo_url":"https://github.com/rpinsler/active-bayesian-coresets","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"active-learning-for-convolutional-neural","repo_url":"https://github.com/svdesai/coreset-al","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"active-learning-for-convolutional-neural","repo_url":"https://github.com/webis-de/small-text","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"active-learning","task_name":"Active Learning"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[{"method_slug":"coresets","method_name":"Coresets"}],"datasets_introduced":[],"methods_introduced":[{"slug":"coresets","name":"Coresets","full_name":"Coresets"}],"results":[{"leaderboard":"/sota/active-learning-on-cifar10-10000","task":"Active Learning","dataset":"CIFAR10 (10,000)","model":"Core-set","rank_in_archive_order":5,"of":7,"metrics":{"Accuracy":"89.92"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1708.00489","atlas_url":"https://app.syntology.ai/?focus=1708.00489","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1708.00489"}},"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/blackhc/active-bayesian-coresets","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/meghshukla/math-analysis-learningloss","reach":{"status":"ok","spdx":"AGPL-3.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/rpinsler/active-bayesian-coresets","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/dsba-lab/openal","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/hillup/active_learning","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/meghshukla/activelearningforhumanpose","reach":{"status":"ok","spdx":"AGPL-3.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/humanlab/rare-class-AL","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/razvancaramalau/Sequential-GCN-for-Active-Learning","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ozansener/active_learning_coreset","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/webis-de/small-text","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/svdesai/coreset-al","reach":null}],"summary":{"ran_violates":1,"ran_fixture":1,"unverified":2},"by_repo_kind":{"listed":{"samples":4,"ran":2,"repositories":2}},"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":4,"samples":[{"code_sha256_prefix":"418af934368adcdb","entry":"cal_distances","repo":"hillup/active_learning","repo_kind":"listed","path":"ac_utils.py","file_url":"https://github.com/hillup/active_learning/blob/HEAD/ac_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"418af934368adcdb"}},{"code_sha256_prefix":"484479ac71cc7fdc","entry":"remove_rows","repo":"svdesai/coreset-al","repo_kind":"listed","path":"active_learn.py","file_url":"https://github.com/svdesai/coreset-al/blob/HEAD/active_learn.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"484479ac71cc7fdc"}},{"code_sha256_prefix":"9f2a097ed541f2e8","entry":"get_features","repo":"svdesai/coreset-al","repo_kind":"listed","path":"active_learn.py","file_url":"https://github.com/svdesai/coreset-al/blob/HEAD/active_learn.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9f2a097ed541f2e8"}},{"code_sha256_prefix":"fd7c9e6b29c78cbd","entry":"get_probs","repo":"svdesai/coreset-al","repo_kind":"listed","path":"active_learn.py","file_url":"https://github.com/svdesai/coreset-al/blob/HEAD/active_learn.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"fd7c9e6b29c78cbd"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}