{"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/unified-spectral-clustering-with-optimal","title":"Unified Spectral Clustering with Optimal Graph","arxiv_id":"1711.04258","date":"2017-11-12","proceeding":null,"authors":["Zhao Kang","Chong Peng","Qiang Cheng","Zenglin Xu"],"abstract":"Spectral clustering has found extensive use in many areas. Most traditional\nspectral clustering algorithms work in three separate steps: similarity graph\nconstruction; continuous labels learning; discretizing the learned labels by\nk-means clustering. Such common practice has two potential flaws, which may\nlead to severe information loss and performance degradation. First, predefined\nsimilarity graph might not be optimal for subsequent clustering. It is\nwell-accepted that similarity graph highly affects the clustering results. To\nthis end, we propose to automatically learn similarity information from data\nand simultaneously consider the constraint that the similarity matrix has exact\nc connected components if there are c clusters. Second, the discrete solution\nmay deviate from the spectral solution since k-means method is well-known as\nsensitive to the initialization of cluster centers. In this work, we transform\nthe candidate solution into a new one that better approximates the discrete\none. Finally, those three subtasks are integrated into a unified framework,\nwith each subtask iteratively boosted by using the results of the others\ntowards an overall optimal solution. It is known that the performance of a\nkernel method is largely determined by the choice of kernels. To tackle this\npractical problem of how to select the most suitable kernel for a particular\ndata set, we further extend our model to incorporate multiple kernel learning\nability. Extensive experiments demonstrate the superiority of our proposed\nmethod as compared to existing clustering approaches.","url_abs":"http://arxiv.org/abs/1711.04258v1","url_pdf":"http://arxiv.org/pdf/1711.04258v1.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":"unified-spectral-clustering-with-optimal","repo_url":"https://github.com/sckangz/AAAI18","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"graph-construction","task_name":"graph construction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1711.04258","atlas_url":"https://app.syntology.ai/?focus=1711.04258","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}