Typically, weights are randomly initialized. Then, as the model is optimized for its given task, those weights are steadily made "better" as determined by the network's loss function. This is also referred to as "training" the neural network.
By far the most popular way of updating weights in a neural net is the backpropagation algorithm, most simply with stochastic gradient descent (SGD). Essentially, the algorithm determines how much each individual weight contributed to the network's loss. It then updates that weight in the direction that would reduce the loss.
I recommend going through Michael Nielsen's online book to learn the basics.