{"ID":2885639,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.04200","arxiv_id":"2508.04200","title":"Bootstrap Deep Spectral Clustering with Optimal Transport","abstract":"Spectral clustering is a leading clustering method. Two of its major shortcomings are the disjoint optimization process and the limited representation capacity. To address these issues, we propose a deep spectral clustering model (named BootSC), which jointly learns all stages of spectral clustering -- affinity matrix construction, spectral embedding, and $k$-means clustering -- using a single network in an end-to-end manner. BootSC leverages effective and efficient optimal-transport-derived supervision to bootstrap the affinity matrix and the cluster assignment matrix. Moreover, a semantically-consistent orthogonal re-parameterization technique is introduced to orthogonalize spectral embeddings, significantly enhancing the discrimination capability. Experimental results indicate that BootSC achieves state-of-the-art clustering performance. For example, it accomplishes a notable 16\\% NMI improvement over the runner-up method on the challenging ImageNet-Dogs dataset. Our code is available at https://github.com/spdj2271/BootSC.","short_abstract":"Spectral clustering is a leading clustering method. Two of its major shortcomings are the disjoint optimization process and the limited representation capacity. To address these issues, we propose a deep spectral clustering model (named BootSC), which jointly learns all stages of spectral clustering -- affinity matrix...","url_abs":"https://arxiv.org/abs/2508.04200","url_pdf":"https://arxiv.org/pdf/2508.04200v1","authors":"[\"Wengang Guo\",\"Wei Ye\",\"Chunchun Chen\",\"Xin Sun\",\"Christian Böhm\",\"Claudia Plant\",\"Susanto Rahardja\"]","published":"2025-08-06T08:30:30Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.LG\"]","methods":"[]","has_code":false,"code_links":[{"ID":611222,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2885639,"paper_url":"https://arxiv.org/abs/2508.04200","paper_title":"Bootstrap Deep Spectral Clustering with Optimal Transport","repo_url":"https://github.com/spdj2271/BootSC","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}