You are the optimizer

You are to act as the optimizer in the standard Gradient Descent training loop.

Your goal is to end up with a set of parameters that minimize the loss.

Once you are done updating, click Done.