198:536 Machine Learning

Spring 2007


Instructor: Dr. Ahmed Elgammal -- email: elgammal at cs

Class Time: Monday 6:40-9:30 pm Hill 254

Office hours: Tuesday 4:30-6:30pm


TA: Nishkam Ravi Email: nravi at cs

TA office hours: Tuesday 4-6pm Core 334


Class Web page: http://www.cs.rutgers.edu/~elgammal/cs536.html


Class Calendar and Lecture Slides




This is a basic graduate-level Machine Learning class that intends to cover a variety of fundamental Machine Learning topics to get you acquainted with the field and its applications.



This is not a comprehensive list of topics nor it reflects the order we will present the topics. Its more a highlevel abstraction of the topics:


Basics: Whats machine learning, dealing with data, Concept Learning, Version Spaces.

Linear Models: Linear Regression, Nave Bayes

Nonlinear Models: Neural Networks, Instant Based Learning, Decision Trees, Boosting, MDL.

Margin-based Approaches: Support Vector Machines and Kernels Methods

Learning theory: VC dimension, PAC learning, Error Bounds

Unsupervised Learning: Clustering, K-means, Expectation Maximization, dimensionality Reduction.

Structured Models: Graphical Models, Hidden Markov Models.

Reinforcement Learning and Evolutionary Learning.

Recommended Background:

Linear algebra and basic probability and statistics.



General Machine Learning text books:


Introduction to Machine Learning, Ethem Alpaydin- MIT press 2004 


Machine Learning, Tom Mitchell. 1997


Pattern Recognition and Machine Learning, Christopher M. Bishop, Springer 2006


Pattern Classification (2nd Edition), Duda, Hart and Stork., Wiley 2001


Topic-Oriented text books:

Learning With Kernels, Bernhard Scholkopf and Alexaner J. Smola

An Introduction To Genetic Algorithms, Melanie Mitchell

Neural Networks for Pattern Recognition, Chris Bishop.

Principle of Data Mining, Hand Mannila, and Smyth

Course Load


Class Project:

Students (in groups of 2 or indviduals) are expected to work on a class research project throughout the semester to explore a recent Machine Learning research topic. Students are to choose their own projects and are encouraged to find a project related to their own research. The project ideas are expected to be innovative, experimental and feasible to be done within the semester time frames.


Project time line (tentative- to be modified)