CS231n课程笔记：Leture4 Backpropagation and Neural Network

目录

How does backpropagation work？

activation functions

---

How does backpropagation work？

 

  

 

we got new value in red  

   another example：

 add gate ：gradient distributor

Q：WHAT'S A MAX GATE? gradient router

the max gate will just take the gradient and route to one of the branches

like this， 

 

 Q：WHAT'S A MUL GATE?

it depends on the another value to multiply

 if it's a vectorized example:

 xT is the highlight

codes for forward/backward API

class ComputationalGraph(object):
    def forward(inputs):
        # 1.[pass inputs to input gates...]
        # 2.forward the computational graph:
        for gate in self.graph.nodes_topologically_sorted():
            gate.forward()
        return loss # the final gate in the graph outputs the loss
    
    def backward():
        for gate in reversed(self.graph.nodes_topologically_sorted()):
            gate.backward()
        return inputs_pradients

codes for multiplygates

class MultiplyGate(object):
    def forward(x, y):
        z = x*y
        return z
    def backward(dz):
        # dx = ... # todo
        # dy = ... # todo
        return [dx, dy]

if you just stack linear layers on top of each other, they're just going to collapse to like a single linear function!!  

activation functions