Universal Function Approximator

less than 1 minute read

Universal Approximation Theorem The universal approximation theorems (UATs) state that neural networks with a certain structure can, in principle, approximate any continuous function to any desired degree of accuracy. Wikipedia Universal Approximation Theorem

Neural Network Visualization
Neural network illustration. Source inspiration: 3Blue1Brown.

🎥 Video Credit “But what is a Neural Network?” by 3Blue1Brown

What are the key missing ingradients for AGI?

You May Also Enjoy

Transformer Architecture Tutorial

less than 1 minute read

Here is the list of good resources to understand transformer architecture. Distilled AI on Transformer Harvard Annotated Transformer ...

Curiosity & Perserverance

less than 1 minute read

This is a good tweet. Riding my first #ebike today and it feels like the future has arrived.— Ben Cichy (@bencichy) November 22, 2019