Statistical and Data Sciences 293 - Modeling for Machine Learning

In the era of “big data,” statistical models are becoming increasingly sophisticated. This course begins with linear regression models and introduces students to a variety of techniques for learning from data, as well as principled methods for assessing and comparing models. Topics include bias-variance trade-off, resampling and cross-validation, linear model selection and regularization, classification and regression trees, bagging, boosting, random forests, support vector machines, generalized additive models, principal component analysis, unsupervised learning and k-means clustering. Emphasis is placed on statistical computing in a high-level language (e.g. R or Python).