Practice — Neural Networks & Backpropagation (6 questions)

A tabular model takes 20 input features, has two hidden layers of 64 ReLU units each, and a single sigmoid output unit for binary classification. All layers are dense with biases.

1. How many learnable parameters does it have? Show the per-layer counts.
2. How much memory do the weights occupy in float32? A colleague says the training memory will be "about the same, since the parameters are all that matter" — what is wrong with that?
3. If you widen the first hidden layer to 128 units (keeping everything else), what is the new parameter count? Does doubling a single layer's width double the model's parameters?