CS 512 - Machine Learning
Fall 2007
|
|
|
Project:
Please note
that the project-2007.doc is updated, and:
- HW data is added to the website (under
project/)
- two more
papers are added, along with local copies of the papers.
Exam:
o 1.5 hr exam on Wednesday 12-1:30 on Dec. 26, not: not regular class
hours!
o questions will be from
the topics covered after the 1st midterm (starting with slide set
6-multivariate)
o number of questions (or
their total) on a subject will be proportional to importance and time-spent on
a subject
o no unnecessary
memorization necessary
Description:
This is a graduate level course
on machine learning. Topics covered will include theoretical aspects of learning (error
bounds, capacity, bias/variance), as well machine basic and advanced machine
learning and pattern recognition algorithms
(Bayesian approaches, Decision trees,NNs,…). The emphasis will be
on what is possible, rather than how to do it (more covered in Pattern
Recognition course). Some of the content will be tailored according to student
composition.
This course is especially
intended for students working in the area of pattern recognition or related
fields, to deepen their understanding of machine learning/pattern recognition
topics. Students who have already taken EE566-Pattern Recognition may find a
significant material being repeated in this course (see the syllabus): while
there will be a significant overlap, this course will cover some topics that were not covered
sufficiently in EE566, particularly theoretical aspects, neural network
approaches and support vector machines.
Prereq: none. Undergraduate level Probability and Linear
Algebra helpful. Matlab or other Toolboxes will be used for homework
assignments.
Book: Machine Learning by
T. Mitchell (ML). Supplementary book: Introduction to Machine
Learning (Ethem), by Ethem Alpaydin. We will follow and cover the ML book
until the end of Chapter 9. We will also read important research articles on
the topic and some student presentations/discussions is
expected. Paper discussion will be at at the end of the term (last 4 weeks).
Course Schedule: Thu: 10:40-12:30 Fri: 3:40-4:30 in
Instructor: Berrin Yanikoglu (berrin@sabanciuniv.edu, FENS 2056)
Office Hours: Walk-in.
Grading: in-class midterm1: %25; in-class
midterm2: %25; in-class final: 30% ; project: 15%; ;
homeworks: 5%
Tentative Syllabus: You may find the slides for the book here. I will
modify them only to some extent.
Week1: 24-28 September
ML1 - Intro to concepts and course - 2
saat
Week 2: 1-5 October
ML2 - Concept Learning
Week 3: 8-12 October (11 & 12th are holidays)
1 hr. of extra course to finish last week’s
material.
Week4: 15-19
October
ML3 - Decision Trees
Week5: 22-26 October
ML4 - ANN (MLP)
Week6: 29 Oct-2 Nov (29th is
a holiday)
ML4 – ANN (readings from Bishop)
ML6 - Bayesian Learning
(Bayes formula, naive Bayes)
Week7: 5-9
November
ML6 – Bayesian
Learning (Bayes formula, naive Bayes) continued
Ethem 4-5:
Intro to multivariate methods
No class on Nov. 9th (TEV meeting)
------------------------------------------------------------------------------------------------------
12-18 November – Break
------------------------------------------------------------------------------------------------------
Week8: 19-23 November
Midterm 1 – 1.5 hrs+ (no class on
the remaining 20 min)
Ethem Chp 5 –
Multivariate methods continued
Week9: 26-30 November
Ethem4: Parameter Estimation (ML, MAP and Bayes estimates)
Slides: Intrinsic error,
Bias-Variance
Week10: 3-7 December
Ethem8,ML8:
Non-parametric density estimation (Parzen windows, KNN)
RBF-Networks
Week 11: 10-14
December
Ethem 7 - Linear Discriminant Analysis (intro)
and Support Vector Machines
Week 12: 17-21
December (20-21:Thu-Fri - Kurban
Bayramı)
No classes.
Week 13: 24-28 December
ML7 - Computational learning Theory (PAC learning, VC dimension)
Mt2: Wednesday 12-1:30
Week 14: 31 December- January
4th (Jan 1st:Tuesday is a holiday)
ML5 - Evaluating hypothesis (Ethem14 - Assesing
& Comparing Class. Algorithms)
Ethem15 - Classifier Combination
Week 15: 7-11 January
Project presentations
Updated Dec. 14, 2007.