{"id":1235,"date":"2021-12-23T05:10:00","date_gmt":"2021-12-22T20:10:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=1235"},"modified":"2021-12-23T05:10:00","modified_gmt":"2021-12-22T20:10:00","slug":"mixtures-of-experts%e3%81%a8%e9%ab%98%e5%af%86%e5%ba%a6%e3%83%a2%e3%83%87%e3%83%ab%e3%81%ae%e6%af%94%e8%bc%83","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=1235","title":{"rendered":"Mixtures of Experts\u3068\u9ad8\u5bc6\u5ea6\u30e2\u30c7\u30eb\u306e\u6bd4\u8f03"},"content":{"rendered":"\n<ul class=\"wp-block-list\"><li><strong>Efficient Large Scale Language Modeling with Mixtures of Experts\u00a0<\/strong>[61.5]<br>\u30a8\u30ad\u30b9\u30d1\u30fc\u30c8\u5c64(MoE)\u306e\u6df7\u5408\u306b\u3088\u308a\u3001\u6761\u4ef6\u4ed8\u304d\u8a08\u7b97\u306b\u3088\u308b\u8a00\u8a9e\u30e2\u30c7\u30eb\u306e\u52b9\u7387\u7684\u306a\u30b9\u30b1\u30fc\u30ea\u30f3\u30b0\u304c\u53ef\u80fd\u306b\u306a\u308b\u3002 \u672c\u7a3f\u3067\u306f, \u81ea\u5df1\u56de\u5e30\u578b MoE \u8a00\u8a9e\u30e2\u30c7\u30eb\u304c, \u5e83\u7bc4\u56f2\u306a\u74b0\u5883\u4e0b\u3067\u306e\u9ad8\u5bc6\u5ea6\u30e2\u30c7\u30eb\u3068\u6bd4\u8f03\u3057\u3066, \u3069\u306e\u3088\u3046\u306b\u30b9\u30b1\u30fc\u30eb\u3059\u308b\u304b\u3092\u793a\u3059\u5b9f\u9a13\u7684\u691c\u8a0e\u3092\u884c\u3063\u305f\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2112.10684v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2112.10684v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Mon, 20 Dec 2021 17:05:11 GMT)<ul><li>\u6700\u8fd1\u3088\u304f\u898b\u308b\u758e\u306a\u30e2\u30c7\u30eb\uff08Mixtures of Experts\uff09\u3068\u5bc6\u306a\u30e2\u30c7\u30eb\u3092\u6bd4\u8f03\u3001\u30bf\u30b9\u30af\u3084\u30c9\u30e1\u30a4\u30f3\u306b\u3082\u3088\u308b\u304c\u758e\u306a\u30e2\u30c7\u30eb\u306f4\u5206\u306e1\u306e\u8a08\u7b97\u91cf\u3067\u5bc6\u306a\u30e2\u30c7\u30eb\u3068\u540c\u7b49\u306e\u6027\u80fd\u3092\u767a\u63ee\u3067\u304d\u308b\u306e\u3067\u306f\uff1f\u3068\u3057\u3066\u3044\u308b\u3002<\/li><li>MoE\u30e2\u30c7\u30eb(1.1T\u30d1\u30e9\u30e1\u30fc\u30bf)\u3068\u8a08\u7b97\u91cf\u7684\u306b\u7b49\u4fa1\u306a\u5bc6\u30e2\u30c7\u30eb(6.7B\u30d1\u30e9\u30e1\u30fc\u30bf)\u306f\u4e00\u8cab\u3057\u3066\u512a\u308c\u3066\u3044\u3066\u3001\u30a8\u30cd\u30eb\u30ae\u30fc\u6d88\u8cbb\u3092\u6e1b\u3089\u3059\u305f\u3081\u306e\u30a2\u30d7\u30ed\u30fc\u30c1\u3068\u3057\u3066\u3082\u6709\u52b9\u3068\u306e\u3053\u3068\u3002<\/li><li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/pytorch\/fairseq\/tree\/main\/examples\/moe_lm\">fairseq\/examples\/moe_lm at main \u00b7 pytorch\/fairseq \u00b7 GitHub<\/a><\/li><\/ul><\/li><\/ul>\n\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Efficient Large Scale Language Modeling with Mixtures of Experts\u00a0[61.5]\u30a8\u30ad\u30b9\u30d1\u30fc\u30c8\u5c64(MoE)\u306e\u6df7\u5408\u306b\u3088\u308a\u3001\u6761\u4ef6\u4ed8\u304d\u8a08\u7b97\u306b\u3088\u308b\u8a00\u8a9e\u30e2\u30c7\u30eb\u306e\u52b9\u7387\u7684\u306a\u30b9 &hellip; <a href=\"https:\/\/devneko.jp\/wordpress\/?p=1235\" class=\"more-link\"><span class=\"screen-reader-text\">&#8220;Mixtures of Experts\u3068\u9ad8\u5bc6\u5ea6\u30e2\u30c7\u30eb\u306e\u6bd4\u8f03&#8221; \u306e<\/span>\u7d9a\u304d\u3092\u8aad\u3080<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[250,567],"class_list":["post-1235","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-mixture-of-experts","tag-567"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/1235","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=1235"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/1235\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=1235"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=1235"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=1235"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}