Problem $10\mathrm{.}17$ What is the receptive field size at each of the first three layers of AlexNet $($i$.$e$.,$ the first three orange blocks in figure $10\mathrm{.}16)?$ Figure $10\mathrm{.}16$ AlexNet $($Krizhevsky et al$.,2012).$ The network maps a $\backslash (224\backslash$times $224\backslash )$ color image to a $1000-$dimensional vector representing class probabilities. The network first convolves with $\backslash (11\backslash$times $11\backslash )$ kernels and stride $4$ to create $96$ channels. It decreases the resolution again using a max pool operation and applies a $\backslash (5\backslash$times $5\backslash )$ convolutional layer. Another max pooling layer follows, and three $\backslash (3\backslash$times $3\backslash )$ convolutional layers are applied. After a final max pooling operation, the result is vectorized and passed through three fully connected $($FC$)$ layers and finally the softmax layer.