Indiana University Bloomington

Informatics I526
Applied Machine Learning

Contact: Sriraam Natarajan
Offered: Fall, Every Year
Capacity: 45
Algebra Required: Basic Algebra
Calculus Required: Basic calculus
Contact Person for Authorization: Sriraam Natarajan
Instructor: Sriraam Natarajan
Days Per Week Offered: Two.
Syllabus: No Syllabus Avaliable
Description: * Introduction to linear regression (and multivariate linear regression)
and practical aspects of implementation
* Logistic Regression and regularization
* Decision trees and pruning, implementation of decision trees
* Support vector machines and making them work in practice
* Boosting - implementing different boosting methods with decision trees.
* Using the algorithms for several tasks - how to set up the problem,
debug, select features and develop the learning algorithm.
* Unsupervised learning - k-means, PCA, hierarchical clustering.
* Implementing the clustering algorithms
* Parallelizing the learning algorithms.
* Applications
* Choosing from multiple algorithms - What will work?
Formal Computing Lab: No
Software: MatLab, C/C++, Python
How Software is Used: Programming Homeworks
Data Analysis: Yes
Presentation: Yes
Exams: Two.