Generally Intelligent cover image

Episode 15: Martín Arjovsky, INRIA, on benchmarks for robustness and geometric information theory

Generally Intelligent

CHAPTER

High Blood Pressure - Cause and Effect

The more data you have for onthe wics, the better you will do. The cause and effect picture is different than the to tisvution picture. In one it bloss a genuine usefulness in ibution performance. And in the other one, it's because of an implicit regularization behavior. This paper lays it out super, superwell. It says that there's two forms of implicit regularization. One is thon by this implicit regulization behavior of s g v. Of that happens of like this thing that i was telling you about before, if you just get to train with no regularization error then you come up with minimum normOne source of this police coeration issue

00:00
Transcript
Play full episode

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
App store bannerPlay store banner