$[$From in$-$class discussion: weeks $4,5]$ Which statement is FALSE regarding how a deep neural network $($model$)$ learns $($aka: deep learning$).$

During training, the goal of a neural network is to minimize error, which is also called the loss or cost function.

Two or more hidden layers are collectively are called a "deep learning algorithm".

Deep learning algorithms propagate the "blame" for an incorrect output $($error$)$ backward through the hidden layers.

During training, the weights in a deep learning algorithm are manually adjusted by an engineer to minimize error.