Transcript Course outline

CS/CMPE 535 –
Machine Learning
Outline
Description

A course on the fundamentals of machine learning –
the science of designing and implementing adaptive
systems
 Concept
learning
 Inductive learning and decision trees
 Computational learning theory
 Bayesian learning
 Graphical models


Emphasis on fundamental mathematical and conceptual
understanding
Significant exposure to real-world implementations and
applications
CS 535 - Machine Learning (Wi 2007-2008) - Asim Karim @ LUMS
2
Goals




To provide a comprehensive introduction to machine
learning methods
To build mathematical foundations of machine
learning and provide an appreciation for its
applications
To provide experience in the implementation and
evaluation of machine learning methods
To develop research interest in the theory and
application of machine learning
CS 535 - Machine Learning (Wi 2007-2008) - Asim Karim @ LUMS
3
Machine Learning is ….



Essential for those who want to specialize in artificial
intelligence and/or want to pursue research in data
mining, machine learning, robotics, computer vision,
and computer networks
Strongly recommended for all graduate students
interested in research
Recommended for students with applied sciences
backgrounds such as engineering
CS 535 - Machine Learning (Wi 2007-2008) - Asim Karim @ LUMS
4
Before Taking This Course…
You should be comfortable with…
 Probability!
 MATH
131 is a prerequisite
 Please revise and keep handy the notes from this course

Artificial intelligence
 General
conceptual understanding would be of much help
 CS331/CS531 is recommended, not required

Programming
 MATLAB
 C/C++
or Java
CS 535 - Machine Learning (Wi 2007-2008) - Asim Karim @ LUMS
5
Grading

Points distribution
Quizzes (~ 7)
Assignments (hand + computer)
Midterm exam
Final exam (comprehensive)
CS 535 - Machine Learning (Wi 2007-2008) - Asim Karim @ LUMS
15%
20%
30%
35%
6
Policies (1)

Quizzes
 Most
quizzes will be announced a day or two in advance
 Unannounced quizzes are also possible

Sharing
 No
copying is allowed for assignments. Discussions are
encouraged; however, you must submit your own work
 Violators can face mark reduction and/or reported to
Disciplinary Committee

Plagiarism
 Do
NOT pass someone else’s work as yours! Write in your
words and cite the reference. This applies to code as well.
CS 535 - Machine Learning (Wi 2007-2008) - Asim Karim @ LUMS
7
Policies (2)

Submission policy
 Submissions
are due at the day and time specified
 Late penalties: 1 day = 10%; 2 day late = 20%; not accepted
after 2 days
 An extension will be granted only if there is a need and when
requested several days in advance.

Classroom behavior
 Maintain classroom sanctity by
remaining quiet and attentive
 If you have a need to talk and gossip, please leave the
classroom so as not to disturb others
 Dozing is allowed provided you do not snore loud 
CS 535 - Machine Learning (Wi 2007-2008) - Asim Karim @ LUMS
8
Policies (3)

Attendance
 Although
attendance is not recorded and graded (in general)
it is strongly recommended. Otherwise, you will miss out on
key understandings not explicitly covered in the textbook
 This recommendation is based on experience of previous
courses
CS 535 - Machine Learning (Wi 2007-2008) - Asim Karim @ LUMS
9
Summarized Course Contents






Introduction, motivation, and applications
Concept learning
Decision tree learning
Evaluating hypotheses and computational learning
theory
Bayesian learning
Graphical models

CS 535 - Machine Learning (Wi 2007-2008) - Asim Karim @ LUMS
10
Course Material

Required textbook


Recommended supplementary text



E. Alpaydin, Introduction to Machine Learning, Pearson
Education, 2004.
C. Bishop, Pattern Recognition and Machine Learning,
Springer, 2006.
Other material


T. Mitchell, Machine Learning, McGraw-Hill, 1997.
Handouts (papers and tutorials as and when necessary)
Other resources


Books in library
Web
CS 535 - Machine Learning (Wi 2007-2008) - Asim Karim @ LUMS
11
Course Web Site

For announcements, lecture slides, handouts,
assignments, quiz solutions, web resources:
http://suraj.lums.edu.pk/~cs535w07/

The resource page has links to information available on
the Web. It is basically a meta-list for finding further
information.
CS 535 - Machine Learning (Wi 2007-2008) - Asim Karim @ LUMS
12
Other Stuff

How to contact me?
 Office
hours: 11.40 to 13.10 MW (office: 429)
 E-mail: [email protected]
 By appointment: to see me outside the office hours e-mail me
for an appointment before coming

Philosophy
 Knowledge
cannot be taught; it is learned.
 Be excited. That is the best way to learn. I cannot teach
everything in class. Develop an inquisitive mind, ask
questions, and go beyond what is required.
 I don’t believe in strict grading. But… there has to be a way
of rewarding performance.
CS 535 - Machine Learning (Wi 2007-2008) - Asim Karim @ LUMS
13
Reference Books in LUMS Library



There are numerous books on machine learning and
related topics in the library.
Browse the library holdings to get a feel of the books
Search the library portal using keywords like “machine
learning”, “learning”, “statistical learning”, etc
CS 535 - Machine Learning (Wi 2007-2008) - Asim Karim @ LUMS
14