This forum is

**inactive**. Browsing/searching possible.
Connect your moderator Slack workspace to receive post notifications:

This forum is **inactive**. Browsing/searching possible.

Connect your moderator Slack workspace to receive post notifications:

## NN Approximation

Hey, it was mentioned in the lecture that for dimension = 3 we need to add an extra layer, but Barron's Approximation states that one hidden NN is enough. So is it enough one hidden layer for d=3 (or more) or not? It would be so helpful if someone can explain this point to me,

Merci,

## 1

## Add comment