Yannic Kilcher Videos (Audio Only) cover image

Neural Networks are Decision Trees (w/ Alexander Mattick)

Yannic Kilcher Videos (Audio Only)

00:00

A Neural Network Is a Fixed Skeleleton With Lots of Blank Spaces

A neural network is a fixed skeleton with lots of blank spaces. The objective of fitting the function in the neural network is figuring out what should be put into its blank spaces this is a parametric approach because we have lots of parameters decision trees are non-parametric approaches so they get optimized along the way. This means it's very hard to fill a decision tree because statistics are always bad and neural network performs way better as it doesn't care about how well it's better but hopes for the best.

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app