{"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/unsupervised-feature-learning-with-c-svddnet","title":"Unsupervised Feature Learning with C-SVDDNet","arxiv_id":"1412.7259","date":"2014-12-23","proceeding":null,"authors":["Dong Wang","Xiaoyang Tan"],"abstract":"In this paper, we investigate the problem of learning feature representation\nfrom unlabeled data using a single-layer K-means network. A K-means network\nmaps the input data into a feature representation by finding the nearest\ncentroid for each input point, which has attracted researchers' great attention\nrecently due to its simplicity, effectiveness, and scalability. However, one\ndrawback of this feature mapping is that it tends to be unreliable when the\ntraining data contains noise. To address this issue, we propose a SVDD based\nfeature learning algorithm that describes the density and distribution of each\ncluster from K-means with an SVDD ball for more robust feature representation.\nFor this purpose, we present a new SVDD algorithm called C-SVDD that centers\nthe SVDD ball towards the mode of local density of each cluster, and we show\nthat the objective of C-SVDD can be solved very efficiently as a linear\nprogramming problem. Additionally, traditional unsupervised feature learning\nmethods usually take an average or sum of local representations to obtain\nglobal representation which ignore spatial relationship among them. To use\nspatial information we propose a global representation with a variant of SIFT\ndescriptor. The architecture is also extended with multiple receptive field\nscales and multiple pooling sizes. Extensive experiments on several popular\nobject recognition benchmarks, such as STL-10, MINST, Holiday and Copydays\nshows that the proposed C-SVDDNet method yields comparable or better\nperformance than that of the previous state of the art methods.","url_abs":"http://arxiv.org/abs/1412.7259v3","url_pdf":"http://arxiv.org/pdf/1412.7259v3.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":[],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"object-recognition","task_name":"Object Recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-classification-on-mnist","task":"Image Classification","dataset":"MNIST","model":"C-SVDDNet","rank_in_archive_order":26,"of":81,"metrics":{"Percentage error":"0.4"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-stl-10","task":"Image Classification","dataset":"STL-10","model":"C-SVDDNet","rank_in_archive_order":95,"of":117,"metrics":{"Percentage correct":"68.2"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}