{"id":920,"date":"2017-11-06T16:59:24","date_gmt":"2017-11-06T14:59:24","guid":{"rendered":"http:\/\/iict-space.heig-vd.ch\/cpn\/?p=920"},"modified":"2017-11-22T11:34:26","modified_gmt":"2017-11-22T09:34:26","slug":"ijcci-2017-international-joint-conference-on-computational-intelligence","status":"publish","type":"post","link":"http:\/\/iict-space.heig-vd.ch\/cpn\/ijcci-2017-international-joint-conference-on-computational-intelligence\/","title":{"rendered":"IJCCI 2017: International Joint Conference on Computational Intelligence"},"content":{"rendered":"

J\u00e9r\u00e9mie Despraz presented at the International Joint Conference on Computational Intelligence<\/h1>\n

Towards a Better Understanding of Deep Neural Networks Representations using Deep Generative Networks<\/span><\/b><\/span><\/span><\/span><\/h3>\n

J\u00e9r\u00e9mie Despraz 1,2<\/sup>, St\u00e9phane Gomez 1,2<\/sup>, H\u00e9ctor F. Satiz\u00e1bal 1<\/sup>, and Carlos Pe\u00f1a-Reyes 1,2<\/sup>
\n1<\/sup> School of Business and Engineering Vaud (HEIG-VD), University of Applied Sciences of Western Switzerland (HES-SO), Yverdon-les-Bains, Switzerland
\n2<\/sup> Computational Intelligence for Computational Biology (CI4CB), SIB Swiss Institute of Bioinformatics, Lausanne, Switzerland<\/span><\/p>\n

\n

Abstract<\/strong>
\nThis paper presents a novel approach to deep-dream-like image generation for convolutional neural networks (CNNs). Images are produced by a deep generative network from a smaller dimensional feature vector. This method allows for the generation of more realistic looking images than traditional activation-maximization methods and gives insight into the CNN\u2019s internal representations. Training is achieved by standard backpropagation algorithms.<\/p>\n<\/div>\n

DOI: 10.5220\/0006495102150222<\/a><\/p>\n

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

J\u00e9r\u00e9mie Despraz presented at the International Joint Conference on Computational Intelligence Towards a Better Understanding of Deep Neural Networks Representations using Deep Generative Networks J\u00e9r\u00e9mie Despraz 1,2, St\u00e9phane Gomez 1,2, H\u00e9ctor F. Satiz\u00e1bal 1, and Carlos Pe\u00f1a-Reyes 1,2 1 School Continue reading IJCCI 2017: International Joint Conference on Computational Intelligence<\/span>→<\/span><\/a><\/p>\n","protected":false},"author":13,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[15],"tags":[],"_links":{"self":[{"href":"http:\/\/iict-space.heig-vd.ch\/cpn\/wp-json\/wp\/v2\/posts\/920"}],"collection":[{"href":"http:\/\/iict-space.heig-vd.ch\/cpn\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/iict-space.heig-vd.ch\/cpn\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/iict-space.heig-vd.ch\/cpn\/wp-json\/wp\/v2\/users\/13"}],"replies":[{"embeddable":true,"href":"http:\/\/iict-space.heig-vd.ch\/cpn\/wp-json\/wp\/v2\/comments?post=920"}],"version-history":[{"count":6,"href":"http:\/\/iict-space.heig-vd.ch\/cpn\/wp-json\/wp\/v2\/posts\/920\/revisions"}],"predecessor-version":[{"id":928,"href":"http:\/\/iict-space.heig-vd.ch\/cpn\/wp-json\/wp\/v2\/posts\/920\/revisions\/928"}],"wp:attachment":[{"href":"http:\/\/iict-space.heig-vd.ch\/cpn\/wp-json\/wp\/v2\/media?parent=920"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/iict-space.heig-vd.ch\/cpn\/wp-json\/wp\/v2\/categories?post=920"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/iict-space.heig-vd.ch\/cpn\/wp-json\/wp\/v2\/tags?post=920"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}