Question: Assume we are a looking at single convolutional layer in a convolutional neural network with a max pooling layer following it as shown below

Assume we are a looking at single convolutional layer in a convolutional neural network with a max pooling

Assume we are a looking at single convolutional layer in a convolutional neural network with a max pooling layer following it as shown below (assume a linear activation function o(x) = x). 110010 000000 001000 1000 10 011100 kernel k 102 0 1 0 3 2 1 convolution with single kemel k (stride 1) (a) Fill in the first two elements of the output of the convolution (the rest is already provided for you). (b) What would be the output of the max pooling layer? Show the 2d matrix. (c) Assume that during back-propagation the derivative of the loss function with respect to the output of the max pooling layer is given below. What is the derivative of the loss function with respect to the output of the convolution layer? Show the 2d matrix. 0.1 0.2 2 350 31 1 2 2x2 Max Pooling (stride 2) 5664 0.3 0.4 (d) Using that result, what is the derivative of the error with respect to woo of the kernel?

Step by Step Solution

3.45 Rating (161 Votes )

There are 3 Steps involved in it

1 Expert Approved Answer
Step: 1 Unlock

Lets address each part of the question using the given information a Fill in the first two elements of the output of the convolution To find these elements we will perform convolution operations with ... View full answer

blur-text-image
Question Has Been Solved by an Expert!

Get step-by-step solutions from verified subject matter experts

Step: 2 Unlock
Step: 3 Unlock

Students Have Also Explored These Related Programming Questions!