In a neural network, inputs

are the constant value given to the weighted input of each node.

determine the speed at which the model can arrive at the most accurate solution.

are variables from the dataset that move information to the next layer via connections.

determine the amount of adjustment made to the weights in the network.