CNN: In this example we try to understand how convolution works in a Convolutional Neural Network (CNN) by manually applying a 3 × 3 filter on a 4 × 4 image matrix shown in the Fig. We also try to...

CNN: In this example we try to understand how convolution works in a Convolutional Neural Network (CNN) by manually applying a 3 × 3 filter on a 4 × 4 image matrix shown in the Fig. We also try to understand forward and
backward pass in CNN by manually calculating for the given image.
1. Consider the input image with only 1 channel in Fig.2. Assume there is no padding for the image, compute the output matrix when the given convolution filter W is applied.
2. After the convolution computation in the above sub problem, a max pool of 2 × 2 is applied. Find the final output ˆy after the pooling layer.