Upon completion of this course a student should be able to:
- Obtain a strong conceptual understanding of statistical learning.
- Learn the statistical principles behind many of the approaches to supervised & unsupervised learning.
- Understand how to perform model selection & evaluation and to effectively communicate the results.
- Learn how to rigorously analyze data using modern statistical methods and computer software.
- Obtain hands-on experience by analyzing real data sets with the skills learned throughout the course.
Evaluation of Students
Students will be graded on written homework assignments, data analysis projects, midterm and final examinations.
The following timeline is approximate.
|Topics||Number of Weeks|
|Introduction to statistical leaning||1 week|
|Linear regression||1 week|
|Methods for model evaluation, model selection and regularization||3 weeks|
|Nonparametric approaches: nearest neighbors, splines, generalized additive models and support vector machine||3 weeks|
|Ensemble methods: bagging, boosting and random forests||2 weeks|
|Unsupervised learning: dimensionality reduction and clustering||2 weeks|
Textbooks & Software
An Introduction to Statistical Learning, with applications in R (2013) by G. James, D. Witten, T. Hastie, R. Tibshirani.
R by the R Development Core Team.
Submitted by: Tao He
Date: May 2, 2016