Firstly:

At end of section 3.5 we are presented with the derivative of the cost function w.r.t. the weights of layer L.

Cost function is defined as C = 0.5(t-y)^2 where t: ‘output from NN’, y:‘target’

In the derivative we are given an equation containing a ‘2’. Can’t see how this value is supossed to be in the equation unless the cost function is C = (t-y)^2

Also; assuming that in layer L, t would be equal to a^L

Secondly:

We are later presented the derivative of the cost function w.r.t the weights of layer L-2. As the continuation of this equation seems to be the derivative w.r.t L-1 rather then L-2, there seems to be an error here.

(The text also talks about L-1 layer, not L-2 layer)

So if my math is wrong, I hope someone will take the time to explain how they arrived at these solutions.

If not are there some editing to for the devs