Episode 16: Yilun Du, MIT, on energy-based models, implicit functions, and modularity
Generally Intelligent
00:00
Is Modularity Important for Deep Learning?
Deep learning was very successful in part because it got rid of a lot of the vadular systems. Iarnin: modularity is very important for performance. Especially, i guess, for reinforcement. In general, i think about these like large onalistic models as big language mottos.
Transcript
Play full episode
