MAB608 Machine Learning


Overview Schedule and Reading Resources

Instructor: Daniel S. Menasche (danielsadoc AT gmail)
Time: Tuesdays and Thursdays 3pm - 5pm
Place: H304A (Coppe Sistemas) [in January, classes will be at Lab 2, Department of Computer Sciece/CCMN/UFRJ [detalhes]
Requisite: undergrad standing


Overview

In the past decade we experienced a dramatic growth in practical applications for machine learning. Yahoo! and Gmail use machine learning to filter spam. Google uses machine learning algorithms to analyze your historic data and predict likely future outcomes. At Ebay, machine learning is used to classify its products and to do contextual advertising. Given the pervasivity of machine learning in our lives, it is not surprising that they have gained the attention of many researchers.

The goal of this course is to provide an introduction to machine learning algorithms. Students will be exposed to machine learning tools such as linear regression, classifiers, neural networks, support vector machines, bayesian networks, hidden Markov models and Markov decision processes. The lectures will emphasize breadth over depth, and will balance practice and theory.


Course Requirements

The class will meet two times a week, for two hours, and will require student participation in several ways,

1) hand in homework assignments

2) solve programming exercises

3) prepare final project or present a paper


Bibliography

The textbook is

C. Bishop, Pattern Recognition and Machine Learning, Springer

Other recommended textbooks:

Duda, Hart, Stork. Pattern Classification (3rd edition).

Cover and Thomas. Elements of Information Theory.