{"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/b-tests-low-variance-kernel-two-sample-tests","title":"B-tests: Low Variance Kernel Two-Sample Tests","arxiv_id":"1307.1954","date":"2013-07-08","proceeding":null,"authors":["Wojciech Zaremba","Arthur Gretton","Matthew Blaschko"],"abstract":"A family of maximum mean discrepancy (MMD) kernel two-sample tests is\nintroduced. Members of the test family are called Block-tests or B-tests, since\nthe test statistic is an average over MMDs computed on subsets of the samples.\nThe choice of block size allows control over the tradeoff between test power\nand computation time. In this respect, the $B$-test family combines favorable\nproperties of previously proposed MMD two-sample tests: B-tests are more\npowerful than a linear time test where blocks are just pairs of samples, yet\nthey are more computationally efficient than a quadratic time test where a\nsingle large block incorporating all the samples is used to compute a\nU-statistic. A further important advantage of the B-tests is their\nasymptotically Normal null distribution: this is by contrast with the\nU-statistic, which is degenerate under the null hypothesis, and for which\nestimates of the null distribution are computationally demanding. Recent\nresults on kernel selection for hypothesis testing transfer seamlessly to the\nB-tests, yielding a means to optimize test power via kernel choice.","url_abs":"http://arxiv.org/abs/1307.1954v3","url_pdf":"http://arxiv.org/pdf/1307.1954v3.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":"b-tests-low-variance-kernel-two-sample-tests","repo_url":"https://github.com/wojzaremba/btest","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"hypothesis-testing","task_name":"Two-sample testing"},{"task_slug":"two","task_name":"Vocal Bursts Valence Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}