Assignment 7: Exploring 3D Sinusoidal Data using Artificial Neural Networks DTSC 680: Applied Machine Learning Name: Directions and Overview The main purpose of this assignment is for you to gain

experience using artificial neural networks to solve simple regression problems. In this assignment, you will fit a neural network to a noisy 3D sinusoidal data set. You will use a Sequential model that can be trained very quickly on the supplied data, so I want you to manually adjust hyperparameter values and observe their influence on the model's predictions. That is, you should manually sweep the hyperparameter space and try to hone in on the reasonable hyperparameter values, again, manually. (Yep, that means guess-and-check: pick some values, train the model, observe the prediction curve, repeat.) So, play around and build some models. When you are done playing with hyperparameter values, you should finish by building an ANN that models the data reasonably well! You should be able to train a model and use it to predict a curve at least as good as mine, but your goal should be to obtain a smoother and less erratic curve. (Side Note: Achieving a less erratic prediction curve could be done either by building a better model, OR by sorting the data more intelligently thereby plotting a prediction curve that looks better. I propose the ideal line is created by sorting the data in such a way that the resulting line minimizes the arc length of the curve. You don't need to worry about any of this, however you do need to generate a figure with a descent-looking prediction curve superimposed on the data.)