Here is my new article on universal approximation theorem
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
Feel free to ask questions or give feedback or both!
1 comment
[ 3.0 ms ] story [ 11.3 ms ] threadHere is my new article on universal approximation theorem
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
Feel free to ask questions or give feedback or both!