In a Fully Connected NN, if the input volume is 32x32x3 connected to a singlelayer of 5 neurons, how many parameters must be learned?

In a fully connected neural network, each neuron in a layer is connected to all neurons in the previous layer. The number of parameters that must be learned is determined by the number of connections between the neurons, plus the bias term for each neuron.

In this case, the input volume is 32x32x3, which means there are 32323 = 3072 neurons in the input layer. This is connected to a layer with 5 neurons.

The number of connections between these layers is therefore 3072 * 5 = 15360.

Each neuron in the second layer also has a bias term, adding 5 more parameters to be learned.

So, the total number of parameters to be learned is 15360 + 5 = 15365.