Processing math: 100%

 

 

 

Using the chain rule and summing over all k entries

We obtain

δlj=kCzl+1kzl+1kzlj=kδl+1kzl+1kzlj,

and recalling that

zl+1j=Mli=1wl+1ijali+bl+1j,

with Ml being the number of nodes in layer l, we obtain

δlj=kδl+1kwl+1kjσ(zlj),

This is our final equation.

We are now ready to set up the algorithm for back propagation and learning the weights and biases.