Solving analytic gradient for multilayer perceptron loss function

[AlanTuring]

TL;DR

Theoretical question concerning the solving of analytic gradient for multilayer perceptron loss function

 

[fresh_42]

This is a very specific question and addresses "Can someone with good theoretical background" only a few users. That is, without any further explanations it will be hard to find someone who is able to answer.

For all others, see e.g. https://scikit-learn.org/stable/modules/neural_networks_supervised.html for context.