tag:blogger.com,1999:blog-7345806147365425073.post5647489932771771014..comments2024-02-29T07:31:17.654+01:00Comments on Peekaboo: [CVML] QuotesAndreas Muellerhttp://www.blogger.com/profile/10177962095394942563noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-7345806147365425073.post-34106796856352253342011-08-02T20:08:09.086+02:002011-08-02T20:08:09.086+02:00I agree, too. But I feel it's less like doing ...I agree, too. But I feel it's less like doing actual magic than summoning some demon and hoping it will do you bidding rather than eating you alive ;)<br /><br />In a totally unrelated note: Hannes has some beautiful new results on segmentation using convolutional nets :)Andreas Muellerhttps://www.blogger.com/profile/10177962095394942563noreply@blogger.comtag:blogger.com,1999:blog-7345806147365425073.post-43553686642687481382011-08-02T13:02:33.975+02:002011-08-02T13:02:33.975+02:00Oh.. I like Malik's third quote: "On grad...Oh.. I like Malik's third quote: "On gradient decent in convolutional neural networks: "This optimization is highly non-convex. But in the hand of a magician like Yann LeCun, it can work."<br /><br />One has to become a magician in order to optimize highly non-convex functions.. I kinda agree ;)Chonoreply@blogger.com