Stat 154: Modern Statistical Prediction and Machine Learning

UC Berkeley

Overview

Theory and practice of statistical prediction. Contemporary methods as extensions of classical methods. Topics: optimal prediction rules, the curse of dimensionality, empirical risk, linear regression and classification, basis expansions, regularization, splines, the bootstrap, model selection, classification and regression trees, boosting, support vector machines. Computational efficiency versus predictive performance. Emphasis on experience with real data and assessing statistical assumptions. This course uses Python as its primary computing language; details are determined by the instructor.

Logistics

Three hours of lecture and two hours of laboratory per week. Four and one-half hours of lecture and three hours of laboratory per week for 10 weeks.

Prerequisites

Mathematics 53 or equivalent; Mathematics 54, Electrical Engineering 16A, Statistics 89A, Mathematics 110 or equivalent linear algebra; Statistics 135, the combination of Data/Stat C140 and Data/Stat/Compsci C100, or equivalent; experience with some programming language. Recommended prerequisite: Mathematics 55 or equivalent exposure to counting arguments.