Computers can learn patterns in data that would be nearly impossible to program manually. Through interactive demos, learn techniques that can detect cancer, identify fraud, and interpret images. Prerequisite: Curiosity! No math background required.
Topics include:
Nearest Neighbor
- Apply nearest neighbor search by hand to classify a data point with numerical features.
- Explain the weaknesses of nearest neighbor search.
Decision Trees
- Apply a given decision tree to classify a given data point.
- Construct a reasonable decision tree by hand given a binary labeled data set.
Logistic Regression
- Construct a good dividing plane by hand, given labeled data points.
- Evaluate the goodness of a given dividing plane visually.
- Given a new data point, use a dividing plane to predict its label.
Gradient Descent
- Identify the local and global minima of a 2 or 3 dimensional curve.
- Illustrate the steps gradient descent would take on such a curve to find the local minimum.
University IT Technology Training classes are only available to Stanford University staff, faculty, or students. A valid SUNet ID is needed in order to enroll in a class.
