{"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/neural-network-based-clustering-using","title":"Neural network-based clustering using pairwise constraints","arxiv_id":"1511.06321","date":"2015-11-19","proceeding":null,"authors":["Yen-Chang Hsu","Zsolt Kira"],"abstract":"This paper presents a neural network-based end-to-end clustering framework.\nWe design a novel strategy to utilize the contrastive criteria for pushing\ndata-forming clusters directly from raw data, in addition to learning a feature\nembedding suitable for such clustering. The network is trained with weak\nlabels, specifically partial pairwise relationships between data instances. The\ncluster assignments and their probabilities are then obtained at the output\nlayer by feed-forwarding the data. The framework has the interesting\ncharacteristic that no cluster centers need to be explicitly specified, thus\nthe resulting cluster distribution is purely data-driven and no distance\nmetrics need to be predefined. The experiments show that the proposed approach\nbeats the conventional two-stage method (feature embedding with k-means) by a\nsignificant margin. It also compares favorably to the performance of the\nstandard cross entropy loss for classification. Robustness analysis also shows\nthat the method is largely insensitive to the number of clusters. Specifically,\nwe show that the number of dominant clusters is close to the true number of\nclusters even when a large k is used for clustering.","url_abs":"http://arxiv.org/abs/1511.06321v5","url_pdf":"http://arxiv.org/pdf/1511.06321v5.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":"neural-network-based-clustering-using","repo_url":"https://github.com/yenchanghsu/NNclustering","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"torch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"neural-network-based-clustering-using","repo_url":"https://github.com/GT-RIPL/L2C","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1511.06321","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}