We're on the edge of a new frontier in art and imagination– and it's not human. Blaise Agüera y Arcas, principal researcher at Google, deals with deep neural networks for machine understanding and distributed knowing. In this captivating demonstration, he shows how neural webs trained to recognize images can be run in reverse, to produce them. The results: magnificent, hallucinatory collages (and poems!) that defy classification. "Understanding and imagination are very totally linked," Agüera y Arcas states. "Any animal, any being that is able to do affective acts is also able to develop."
TEDTalks is a daily video podcast of the best talks and performances from the TED Conference, where the world's leading thinkers and doers offer the talk of their lives in 18 minutes (or less). Try to find talks on Technology, Entertainment and Design– plus science, business, global concerns, the arts and far more.
Find closed captions and equated subtitles in lots of languages at
Follow TED news on Twitter:
Like TED on Facebook:
Register for our channel:
Leave a Reply