CS6140 Machine Learning, SPRING 2015

Announcements (please check this page frequently)


Class Times:

Monday, 6:00-9:00, 320 Behrakis Health Sciences Center

Instructor:

Teaching Assistants:

What is this course about?

Machine learning is primarily concerned with collecting and analyzing data. Machine learning algorithms focus on eliciting hidden patterns and trends in data that can be used for making predictions and intelligent decisions. There are a number of domains in which machine learning techniques have been applied successfully, these include medical imaging, clinical diagnosis, finance, marketing, etc. Widespread use and commerical success has led to an increased interest in machine learning.
The main aim of this course is to introduce machine learning to graduate students. Our main emphasis will be on supervised learning methods, along with an introduction to their underlying mathemtical models. Therefore, some mathematical aptitude is required especially basic knowledge of Linear Algebra and Probability theory, lectures will cover the necessary background on any mathematical concept/tehcnique that is required for understanding the content.

Syllabus:

Topics include supervised learning methods for classification and regression and unsupervised learning for clustering data. For most of the learning algorithms we will explore different learning frameworks such as maximum likelihood and Bayesian statistics.
List of Topics (Note: We may not cover all sub-topics)

Prerequisites:

Most of the background concepts will be covered during class to serve as a "refresher", and clarify the lecture content.

Course Work and Grading

The course grade will be determined solely only on the bi-weekly assignments. Each assignment has a programming component and a written component, with the following grading breakdown for each assignment:
Written problems (20%)
these are relatively short problems which reinforce the class material
Experimental/Programming problems (80%)
these problem will either require you to implement a learning algorithm in your choice of programming language and then evaluate its performance on publicly available datasets (these will be provided through the course scedule page). These problems will be graded by the instructor or the TA during the help sessions where you will be required to give a demo of your code in person.
Rules for late submissions:
Homework is due at the beginning of class on the announced due date. You will be granted one homework extension of 1 week, to be used at your discretion, no questions asked. After the first late assignment, unexcused late assignments will be penalized 20% per calendar day late . Submissions after the date on which the following assignment is due or after the solutions have been handed out will not be accepted. If you have a valid reason for turning in a late assignment, please see/email me in advance
A note on assignments:
We expect the average total load to be 20 hours/week. Based on past experience, many students spend more than 20 hours/week for additional work denoted "Extra Credit"; the EC can improve your grade, but it is not a replacement for regular credit; EC is harder, more time consuming, worth less points, and less discussed during helping sessions; it is designed for students who want to get more out of the course. The best way, by far, to maximize your grade is to complete the regular credit on time.
Submission Instructions:

All assignments have a programming component and a written component.

Submitting Programming Solutions:

Please submit all your code along with a README.txt file that has clear and specific instructions/steps on:

  1. How to compile your code.
  2. How to run your code.

We should be able to compile and run your code on the lab machines at CCIS.

Submitting Written Solutions:

Submit your solutions to the theory/written problems in the form of a PDF file. You can scan your handwritten solutions and create a PDF.

Demonstration of Programming Assignments:

The final grade for the programming problems will be posted after a demonstration has been provided for the submission during course office hours to either the TA or the instructor. The demonstration should be provided no later than one week after the submission deadline.

Collaboration:

On homework assignments and projects: You may discuss the problems and general ideas about their solutions with other students, and similarly you may consult other textbooks or the web. However, you must work out the details on your own and code/write-out the solution on your own. Every such collaboration (either getting help or providing help) and every use of text or electronic sources must be clearly cited and acknowledged in the submitted homework. Failure to follow these guidelines may result in disciplinary action for all parties involved. Any questions for this and other issues concerning academic integrity please consult the policies regarding academic integrity available from the office of student conduct and conflict resolution (OSCRR).

Textbooks and Material Covered

The class materials are based on a number of different texts, therefore there is no singe textbook for this course. For each lecture the schedule will list the recommended/relevant readings from the following textbooks: All class notes and lecture slides along with links to supplementary materials will be regularly posted on the schedule page.