Hello everyone<p>Here is my new article on universal approximation theorem<p>TLDR:
- I implement a first universal approximator with TesnorFlow and i train it on a sinus function (I show that it actually works)
- I use it inside a bigger neural networks to classify the MNIST dataset
- I display the learnt activation functions
- I show that whatever the learnt activation function is, i get consistently the accuracy 0.98 on the test set
- bonus: all the code is open-source<p>Feel free to ask questions or give feedback or both!