
Episode 16: Yilun Du, MIT, on energy-based models, implicit functions, and modularity
Generally Intelligent
00:00
Are You Trying to Beat the State of the Art?
Most of your work does not seem to be about making the prettiest pictures, or like beating the state of the art by the very most. Do you just kind of like, play around with it and then like, oh, great, we'll just find something that seems good enough that this can be published? It almost sounds like you're doing the former, the likolders playing around, and then we'll do whatever we need to do to get a publish. Is that fair? Yes, it's kind of true. So deffntly, like none of my research is really stated art. You could say, like we compare with some base lines, afd base lines
Transcript
Play full episode