Assigned Wednesday, October 8
Due Friday, October 17
In this mini-homework, you are to develop a new hypothesis based on the 20 labeled training examples you created in the first mini-homework. This hypothesis may be a simple linear threshold unit, a multi-layer neural network, or a support vector machine. You may craft your hypothesis manually or use one of the learning algorithms discussed in class.
In your writeup, include some graphical representation of your hypothesis, and explain why it is designed the way it is, e.g., why you chose your particular architecture or kernel and why the weights are the way they are.
Take your writeup and graphical representation and add it to your description of your target concept, attributes, examples with labels and attribute values, and decicion tree in your personal wiki, which is already linked in from the course wiki. Do not remove your previous results from your wiki; keep everything as a single, comprehensive document.