Indiana University Bloomington

Statistics S682
Topics In Machine Learning Ii

Offered: Spring, 2014
Capacity: 30
Pre-Requisites: S681 Topics in Machine Learning I or STAT-S722 or equivalent or Permission of instructor
Algebra Required: Yes
Calculus Required: Yes
Contact Person for Authorization: Daniel McDonald
Instructor: Daniel McDonald
Days Per Week Offered: 2
Website: http://mypage.iu.edu/~dajmcdon/
Recommended follow-up classes: STAT-S781, STAT-S722-722, STAT-S675
Syllabus: No Syllabus Avaliable
Keywords: Machine learning, data mining, big data, statistical learning, regression, classification, nonparametrics, optimization
Description: This two-semester course sequence emphasizes supervised machine learning techniques.
Strong emphasis is given to the theoretical and statistical aspects of the methods covered. In addition, practical aspects of methodology, the underlying concepts and intuition behind each method, and computational considerations are addressed. The goal is to provide a balance between the "art" of designing good learning algorithms and the "science" of analyzing an algorithm's statistical properties and performance guarantees.

The sequence will provide students with a deep understanding of the blend of methodological and computational concerns with statistical theoretic analysis. It will also allow students to select methods appropriate to problems in their own research.
Books: Hastie, Tibshirani, and Friedman The Elements of Statistical Learning: Data Mining, Inference, and Prediction
Substantive Orientation: The emphasis is on statistical methodology as it applies broadly
Statistical Orientation: Methodology and theory with example applications
Applied/Theoretical: The focus is on various statistical techniques and some of the theory used to justify their application
Formal Computing Lab: No
Software: R
How Software is Used: For homework sets and projects
Problem Sets: Yes
Data Analysis: Yes
Presentation: Yes
Exams: Yes