{"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/breaking-the-reclustering-barrier-in-centroid","title":"Breaking the Reclustering Barrier in Centroid-based Deep Clustering","arxiv_id":"2411.02275","date":"2024-11-04","proceeding":null,"authors":["Lukas Miklautz","Timo Klein","Kevin Sidak","Collin Leiber","Thomas Lang","Andrii Shkabrii","Sebastian Tschiatschek","Claudia Plant"],"abstract":"This work investigates an important phenomenon in centroid-based deep clustering (DC) algorithms: Performance quickly saturates after a period of rapid early gains. Practitioners commonly address early saturation with periodic reclustering, which we demonstrate to be insufficient to address performance plateaus. We call this phenomenon the \"reclustering barrier\" and empirically show when the reclustering barrier occurs, what its underlying mechanisms are, and how it is possible to Break the Reclustering Barrier with our algorithm BRB. BRB avoids early over-commitment to initial clusterings and enables continuous adaptation to reinitialized clustering targets while remaining conceptually simple. Applying our algorithm to widely-used centroid-based DC algorithms, we show that (1) BRB consistently improves performance across a wide range of clustering benchmarks, (2) BRB enables training from scratch, and (3) BRB performs competitively against state-of-the-art DC algorithms when combined with a contrastive loss. We release our code and pre-trained models at https://github.com/Probabilistic-and-Interactive-ML/breaking-the-reclustering-barrier .","url_abs":"https://arxiv.org/abs/2411.02275v1","url_pdf":"https://arxiv.org/pdf/2411.02275v1.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":"breaking-the-reclustering-barrier-in-centroid","repo_url":"https://github.com/probabilistic-and-interactive-ml/breaking-the-reclustering-barrier","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"deep-clustering","task_name":"Deep Clustering"},{"task_slug":"image-clustering","task_name":"Image Clustering"},{"task_slug":"unsupervised-image-classification","task_name":"Unsupervised Image Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-clustering-on-cifar-10","task":"Image Clustering","dataset":"CIFAR-10","model":"DCN+BRB","rank_in_archive_order":10,"of":40,"metrics":{"ARI":"0.824","Accuracy":"0.912","Backbone":"ResNet-18","NMI":"0.837","Train set":"Train"},"uses_additional_data":false},{"leaderboard":"/sota/image-clustering-on-cifar-10","task":"Image Clustering","dataset":"CIFAR-10","model":"IDEC+BRB","rank_in_archive_order":11,"of":40,"metrics":{"ARI":"0.818","Accuracy":"0.907","Backbone":"ResNet-18","NMI":"0.833","Train set":"Train"},"uses_additional_data":false},{"leaderboard":"/sota/image-clustering-on-cifar-10","task":"Image Clustering","dataset":"CIFAR-10","model":"DEC+BRB","rank_in_archive_order":12,"of":40,"metrics":{"ARI":"0.812","Accuracy":"0.906","Backbone":"ResNet-18","NMI":"0.826","Train set":"Train"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-image-classification-on-cifar-20","task":"Unsupervised Image Classification","dataset":"CIFAR-20","model":"DCN+BRB","rank_in_archive_order":4,"of":14,"metrics":{"ARI":"41.15","Accuracy":"56.92","NMI":"56.76"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-image-classification-on-cifar-20","task":"Unsupervised Image Classification","dataset":"CIFAR-20","model":"IDEC+BRB","rank_in_archive_order":7,"of":14,"metrics":{"ARI":"38.81","Accuracy":"55.43","NMI":"54.81"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-image-classification-on-cifar-20","task":"Unsupervised Image Classification","dataset":"CIFAR-20","model":"DEC+BRB","rank_in_archive_order":11,"of":14,"metrics":{"ARI":"35.05","Accuracy":"50.46","NMI":"51.72"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2411.02275","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2411.02275"}},"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/probabilistic-and-interactive-ml/breaking-the-reclustering-barrier","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"summary":{"unverified":5},"by_repo_kind":{"official":{"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":"13639c9df329a5ae","entry":"adam_reset_centers_momentum","repo":"probabilistic-and-interactive-ml/breaking-the-reclustering-barrier","repo_kind":"official","path":"src/deep/brb_reclustering.py","file_url":"https://github.com/probabilistic-and-interactive-ml/breaking-the-reclustering-barrier/blob/HEAD/src/deep/brb_reclustering.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"13639c9df329a5ae"}},{"code_sha256_prefix":"7bc460bf821933f4","entry":"get_normalize_fn","repo":"probabilistic-and-interactive-ml/breaking-the-reclustering-barrier","repo_kind":"official","path":"src/datasets/augmentation_dataloader.py","file_url":"https://github.com/probabilistic-and-interactive-ml/breaking-the-reclustering-barrier/blob/HEAD/src/datasets/augmentation_dataloader.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"7bc460bf821933f4"}},{"code_sha256_prefix":"adc59ade92bb9af0","entry":"get_run_information","repo":"probabilistic-and-interactive-ml/breaking-the-reclustering-barrier","repo_kind":"official","path":"wandb_downloader.py","file_url":"https://github.com/probabilistic-and-interactive-ml/breaking-the-reclustering-barrier/blob/HEAD/wandb_downloader.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"adc59ade92bb9af0"}},{"code_sha256_prefix":"e432ab3020785a30","entry":"load_configs_from_files","repo":"probabilistic-and-interactive-ml/breaking-the-reclustering-barrier","repo_kind":"official","path":"runner.py","file_url":"https://github.com/probabilistic-and-interactive-ml/breaking-the-reclustering-barrier/blob/HEAD/runner.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"e432ab3020785a30"}},{"code_sha256_prefix":"8bb08e6cebcb6d20","entry":"update_centers_with_labels","repo":"probabilistic-and-interactive-ml/breaking-the-reclustering-barrier","repo_kind":"official","path":"src/deep/brb_reclustering.py","file_url":"https://github.com/probabilistic-and-interactive-ml/breaking-the-reclustering-barrier/blob/HEAD/src/deep/brb_reclustering.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"8bb08e6cebcb6d20"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}