Yannic Kilcher Videos (Audio Only) cover image

Neural Networks are Decision Trees (w/ Alexander Mattick)

Yannic Kilcher Videos (Audio Only)

CHAPTER

Neural Networks and Decision Trees Using a Theorem

When neural networks are large and just basically large enough to fit everything inside of them that means that the actual size of these neural network trees can become rather gigantic. The way we can do analysis with a theoretical lens is by studying something called the VC dimension or the bub mixture one end-const dimension which effectively tells us how many different points can network distinguish. For example for a decision tree if you have a fully balanced tree like this one would be two to the power of the depth of the tree while for a neural network it's a lot harder to figure out because you have all of these different architectures.

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