Problem Set 1
1. Artificial Intelligence versus Machine Learning versus Deep Learning [6 points]: In your own words, define artificial intelligence, define machine learning, and define deep learning. Then describe the relationship
between the three disciplines.
2. Supervised Learning Generalization [4 points]: Describe the motivation for splitting data into a training dataset and test dataset and the
purpose of each subset for developing deep learning algorithms.
3. Artificial Neurons:
(a) Model Training [6 points]: Show the mathematical steps of learning
a Perceptron model over two epochs, using the training data shown in
Table 1, model weights initialized to 0, and learning rate of 0.1. For
full credit, you must include the mathematical steps used to derive the
weights and two tables showing resulting weights after each training
update round (i.e., after each training example) for each epoch. (This
is not a programming exercise)
Sample X1 X2 X3 Y
Training
1 0 0 0 1
2 1 0 0 1
3 1 1 0 -1
Test
1 1 1 0 -1
2 1 0 1 -1
3 1 1 1 1
4 0 0 0 1
Table 1: Training and test datasets.
1
(b) Model Testing [3 points]: Report prediction results for the test data
shown in Table 1.
(c) Model Evaluation [3 points]: Evaluate the prediction results by
showing the confusion matrix.
(d) Model Evaluation [3 points]: Evaluate the prediction results by
reporting the precision and recall.
2
