Syllabus: Introduction to Machine Learning (CS189/289A)

(approximate)

• Introduction: applications, methods, concepts
• Good Machine Learning hygiene: test/training/validation, overfitting
• Linear classification
• Perceptron algorithm
• Support vector machines (SVMs)
• Statistical learning background
• Decision theory; Bayes risk
• Probabilistic models vs no model
• Generative and discriminative models
• Controlling complexity: regularization, bias-variance trade-off, priors.
• Resampling, cross-validation.
• The multivariate normal distribution.
• Linear regression
• Least squares
• Regularization: ridge regression, lasso
• Brief primer on optimization
• Linear Classification, revisited
• Logistic regression
• Linear Discriminant Analysis
• Support vector machines revisited
• Algorithms
• The kernel trick
• Theoretical analysis of machine learning problems and algorithms
• Generalization error bounds; VC dimension
• Nearest neighbor methods
• k-nearest-neighbor
• Properties of high-dimensional spaces
• Distance learning
• Efficient indexing and retrieval methods
• Decision trees
• Classification and regression trees
• Random Forests
• Boosting
• Neural networks
• Multilayer perceptrons
• Variations such as convolutional nets
• Applications
• Unsupervised methods
• Clustering
• Density estimation
• Dimensionality reduction
• Applications in Data Mining
• Collaborative filtering
• The power and the peril of Big Data