{"id":1950,"date":"2015-06-22T13:21:25","date_gmt":"2015-06-22T11:21:25","guid":{"rendered":"http:\/\/www.zerga.de\/wordpress\/?p=1950"},"modified":"2015-06-22T13:21:25","modified_gmt":"2015-06-22T11:21:25","slug":"neural-network-dream","status":"publish","type":"post","link":"https:\/\/zerga.de\/wordpress\/2015\/06\/22\/neural-network-dream\/","title":{"rendered":"What does a neural network dream of?"},"content":{"rendered":"<div style=\"width: 510px\" class=\"wp-caption aligncenter\"><a href=\"http:\/\/images.zeit.de\/digital\/2015-06\/google-fs-bilder\/13-google.jpg\" rel=\"lightbox[1950]\"><img loading=\"lazy\" class=\"\" src=\"http:\/\/images.zeit.de\/digital\/2015-06\/google-fs-bilder\/13-google.jpg\" alt=\"\" width=\"500\" height=\"281\" \/><\/a><p class=\"wp-caption-text\">&#8220;Was immer du hier siehst, ver\u00e4ndere das Bild so, dass es st\u00e4rker wird!&#8221; lautete die Anweisung der Google-Forscher an das neuronale Netzwerk. Und so fing das Netzwerk an, die Formen der Wolken auf dem Bild zu interpretieren: Was etwa ein bisschen nach Vogel aussah, wurde noch vogel\u00e4hnlicher gemacht. So &#8220;malte&#8221; das k\u00fcnstliche Gehirn allerlei Fantasiewesen und Geb\u00e4ude, die es zu erkennen glaubte. | Zeit.de &#8211; \u00a9 Google Research<\/p><\/div>\n<p>Google made multiple layers of neural networks analyze images and emphasize what they &#8220;think&#8221; they found in the pictures. The result is really impressive even though some of the pictures they were fed contained nothing but white noise to begin with. One could argue if that&#8217;s how AI dreams are made&#8230; well, and maybe even our own?<\/p>\n<p style=\"text-align: center\"><a href=\"http:\/\/www.zeit.de\/digital\/2015-06\/google-neuronale-netzwerke-fs\" target=\"_blank\">http:\/\/www.zeit.de\/digital\/2015-06\/google-neuronale-netzwerke-fs<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Google made multiple layers of neural networks analyze images and emphasize what they &#8220;think&#8221; they found in the pictures. The result is really impressive even though some of the pictures they were fed contained nothing but white noise to begin with. One could argue if that&#8217;s how AI dreams are made&#8230; well, and maybe even [&#038;hellip<\/p>\n","protected":false},"author":2,"featured_media":1953,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ngg_post_thumbnail":0},"categories":[17,6],"tags":[390,295,392,394,64,105,393,32,391],"_links":{"self":[{"href":"https:\/\/zerga.de\/wordpress\/wp-json\/wp\/v2\/posts\/1950"}],"collection":[{"href":"https:\/\/zerga.de\/wordpress\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/zerga.de\/wordpress\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/zerga.de\/wordpress\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/zerga.de\/wordpress\/wp-json\/wp\/v2\/comments?post=1950"}],"version-history":[{"count":5,"href":"https:\/\/zerga.de\/wordpress\/wp-json\/wp\/v2\/posts\/1950\/revisions"}],"predecessor-version":[{"id":1957,"href":"https:\/\/zerga.de\/wordpress\/wp-json\/wp\/v2\/posts\/1950\/revisions\/1957"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/zerga.de\/wordpress\/wp-json\/wp\/v2\/media\/1953"}],"wp:attachment":[{"href":"https:\/\/zerga.de\/wordpress\/wp-json\/wp\/v2\/media?parent=1950"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/zerga.de\/wordpress\/wp-json\/wp\/v2\/categories?post=1950"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/zerga.de\/wordpress\/wp-json\/wp\/v2\/tags?post=1950"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}