CSE204 - Machine Learning

Information

Course outline: a working outline of topics (non-exhaustive):

• Introduction to Machine Learning
• Regression (least squares, polynomial regression, gradient descent, k-nearest neighbors)
• Classification (logistic regression, k-nearest neighbors)
• Overfitting and regularization (ridge regression)
• Neural Networks I
• Neural Networks II (multi-layer perceptrons and back propagation)
• Deep Learning (and convolutional neural networks)
• Decision trees and ensemble methods (C4.5, bagging, boosting, random forest)
• Unsupervised Learning I (principal components analysis, autoencoders)
• Unsupervised Learning II (k-means clustering, mixture models)
• Kernel methods (support vector machines, spectral clustering)
• Reinforcement learning

Grading

• Lab reports + quizzes: 50% (two in-class lab exams, one graded assignment, and minor points for lab completions)
• Group project (in groups of 3): 50% (evaluated via an approx. 30-minute oral presentation at the end of the course)

Recommended readings

(my picks)

• Good book to review fundamentals: Mathematics for Machine Learning (Deisenroth et al.)
• Focus on Linear Algebra: Linear Algebra and Learning from Data (Gilbert Strang)
• The “bible” of Deep Learning: Deep Learning (Goodfellow et al.)