
Episode 16: Yilun Du, MIT, on energy-based models, implicit functions, and modularity
Generally Intelligent
00:00
Is There a Future Work at Ning Be Excited About the Paper?
In the real world, objects are ke in the same spot for a long time. And over a video, that's very obvious. But in a picture, it's kind of hard to tell what's associated with what,. Like an information theoretic sense. It also seems like it might be even easier for things to be disentangled in that sense. I think another imforni is folly. Video, and i think frothe you ask, will probably say that ing to discover objects and images a bad ideaieyou really want to do things inoson tevideo a.
Transcript
Play full episode