{"id":813,"date":"2024-10-06T13:13:31","date_gmt":"2024-10-06T13:13:31","guid":{"rendered":"https:\/\/blue16.cn\/?p=813"},"modified":"2024-10-11T05:02:46","modified_gmt":"2024-10-11T05:02:46","slug":"%e6%94%af%e6%8c%81%e7%9f%a9%e9%98%b5%e6%9c%ba","status":"publish","type":"post","link":"https:\/\/blue16.cn\/index.php\/2024\/10\/06\/813\/","title":{"rendered":"\u652f\u6301\u77e9\u9635\u673a"},"content":{"rendered":"\n

\u95ee\u9898\u80cc\u666f<\/h2>\n\n\n\n

\u8111\u7535\u56fe\u3001\u56fe\u50cf\u5206\u7c7b\uff1a\u8f93\u5165\u7684\u4fe1\u606f\u662f\u77e9\u9635\u3002\u4f20\u7edf\u7684\u652f\u6301\u5411\u91cf\u673a\uff1a\u4f1a\u5ffd\u7565\u7ed3\u6784\u4fe1\u606f\u3002<\/p>\n\n\n\n

\u4e3a\u4e86\u5229\u7528\u8fd9\u79cd\u7ed3\u6784\u4fe1\u606f\uff0c\u6211\u4eec\u63d0\u51fa\u4e00\u79cd\u65b0\u7684\u5206\u7c7b\u65b9\u6cd5\uff0c\u6211\u4eec\u79f0\u4e4b\u4e3a\u652f\u6301\u77e9\u9635\u673a(SMM)<\/p>\n\n\n\n

SMM = Hinge Loss + spectral elastic net penalty<\/p>\n\n\n\n

SMM\u95ee\u9898\uff1a\u51f8\u4f18\u5316->\u91c7\u7528ADMM\u7b97\u6cd5<\/p>\n\n\n\n

\u4ecb\u7ecd<\/h2>\n\n\n\n

\u6838\u5fc3\uff1a\u8bd5\u56fe\u5229\u7528\u6570\u636e\u77e9\u9635\u7684\u5217\u6216\u884c\u4e4b\u95f4\u7684\u76f8\u5173\u6027\uff08\u5229\u7528\u7ed3\u6784\u4fe1\u606f\uff09\u3002<\/p>\n\n\n\n

\n

Wolf et al (2007) proposed a rank-k SVM, which models the regression matrix as the sum of k rank-one orthogonal matrices. Pirsiavash et al (2009) devised a bilinear SVM by factorizing the regression matrix into two lowrank matrices. Cai et al (2006) proposed a similar bilinear framework called support tensor machines for text categorization.<\/p>\n<\/blockquote>\n\n\n\n

\u4e4b\u524d\u7684\u7814\u7a76\uff1arank-k SVM\u3001<\/p>\n\n\n\n

\u5bfc\u81f4\u7684\u95ee\u9898\uff1a\u975e\u51f8\u4f18\u5316\u95ee\u9898<\/p>\n\n\n\n

\u672c\u7bc7\u8bba\u6587\u7684\u6a21\u578b\u7684\u521b\u65b0\u70b9\uff1a<\/p>\n\n\n\n

\n

First, we consider the hinge loss due to its widely deployed ability in sparseness and robustness modeling. Second, we employ a spectral elastic net penalty for the regression matrix.<\/p>\n<\/blockquote>\n\n\n\n

    \n
  • \u635f\u5931\u51fd\u6570\uff1ahinge loss<\/li>\n\n\n\n
  • \u60e9\u7f5a\u9879\uff1aspectral elastic net penalty\uff08\u4f20\u7edf\u7684\uff1aelastic net\uff09<\/li>\n<\/ul>\n\n\n\n

    \u4e09\u79cd\u8303\u6570<\/h3>\n\n\n\n

    \u5f17\u7f57\u8d1d\u5c3c\u4e4c\u65af\u8303\u6570\uff08Frobenius norm\uff09<\/h4>\n\n\n\n

    \u6709\u70b9\u7c7b\u4f3c\u4e8e\u6211\u4eec\u5e38\u89c1\u7684\u8ddd\u79bb\u7684\u8ba1\u7b97\u65b9\u6cd5\uff1a\u5e73\u65b9\u5f00\u6839\u53f7<\/p>\n\n\n

    \n
    \"\"<\/div><\/figure>\n<\/div>\n\n\n

    \u6838\u8303\u6570\uff08Nuclear norm\uff09<\/h4>\n\n\n\n

    \u5c31\u662f\u77e9\u9635\u5728\u5947\u5f02\u503c\u5206\u89e3\uff08SVD\uff09\u4e2d\u77e9\u9635\u03a3\u5bf9\u89d2\u7ebf\u4e0a\u6240\u6709\u5947\u5f02\u503c\u7684\u548c\uff1a<\/p>\n\n\n

    \n
    \"\"<\/div><\/figure>\n<\/div>\n\n\n

    \u8c31\u8303\u6570\uff08Spectral norm\uff09<\/h4>\n\n\n\n

    \u77e9\u9635\u7684\u6700\u5927\u5947\u5f02\u503c\u3002<\/p>\n\n\n

    \n
    \"\"<\/div><\/figure>\n<\/div>\n\n
    \n
    \"\"<\/div><\/figure>\n<\/div>\n\n\n

    \u652f\u6301\u77e9\u9635\u673a(SMM)<\/h2>\n\n\n
    \n
    \"\"<\/div><\/figure>\n<\/div>\n\n\n

    <\/p>\n","protected":false},"excerpt":{"rendered":"

    \u95ee\u9898\u80cc\u666f \u8111\u7535\u56fe\u3001\u56fe\u50cf\u5206\u7c7b\uff1a\u8f93\u5165\u7684\u4fe1\u606f\u662f\u77e9\u9635\u3002\u4f20\u7edf\u7684\u652f\u6301\u5411\u91cf\u673a\uff1a\u4f1a\u5ffd\u7565\u7ed3\u6784\u4fe1\u606f\u3002 \u4e3a\u4e86\u5229\u7528\u8fd9\u79cd\u7ed3\u6784\u4fe1\u606f\uff0c\u6211\u4eec\u63d0 […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[26,47,46,45],"tags":[],"class_list":["post-813","post","type-post","status-publish","format-standard","hentry","category-study","category-47","category-46","category-45"],"_links":{"self":[{"href":"https:\/\/blue16.cn\/index.php\/wp-json\/wp\/v2\/posts\/813","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/blue16.cn\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blue16.cn\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blue16.cn\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/blue16.cn\/index.php\/wp-json\/wp\/v2\/comments?post=813"}],"version-history":[{"count":3,"href":"https:\/\/blue16.cn\/index.php\/wp-json\/wp\/v2\/posts\/813\/revisions"}],"predecessor-version":[{"id":943,"href":"https:\/\/blue16.cn\/index.php\/wp-json\/wp\/v2\/posts\/813\/revisions\/943"}],"wp:attachment":[{"href":"https:\/\/blue16.cn\/index.php\/wp-json\/wp\/v2\/media?parent=813"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blue16.cn\/index.php\/wp-json\/wp\/v2\/categories?post=813"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blue16.cn\/index.php\/wp-json\/wp\/v2\/tags?post=813"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}