Neural Networks are universal function approximators.

[KBleivik]

Sorry for being a little theoretic in this thread. I could post it on our OOPShool forum. Yes, I wrote about neural networks in my cand scient thesis in mathematics at the Univdersity of Oslo in 1996 (I have a reference if you are serious and ask) and I have an academic degree in Economics and education at the Doctor Scient Level in Numerical analysis.

So here is the video I want to share with you today:

Watching Neural Networks Learn

Semantic related informaion:

The Universal Approximation Theorem

Michael Nielsen | The Universal Approximation Theorem for neural networks

Computational Thinking | The Universal Approximation Theorem of Neural Networks

Scientific articles about Neural Networks as universal function approximators.

Will AI Take Over the Jobs? - Jack Ma Speaks on AI Vs Machine Learning Vs Data Science

But what is a neural network? | Chapter 1, Deep learning

Why Neural Networks can learn (almost) anything

3Blue1Brown, animated math | some resources for additional information:

https://www.3blue1brown.com/

https://www.youtube.com/@3blue1brown

https://www.patreon.com/3blue1brown

https://twitter.com/3blue1brown

https://www.reddit.com/r/3Blue1Brown/

https://www.instagram.com/3blue1brown

https://www.facebook.com/3blue1brown New video every day with probability of 0.05.

https://store.dftba.com/collections/3blue1brown