CSC 462: Machine Learning
Schedule and Office Hours
- Classes: Monday at 13:00 - 14:50 and Wednesday at 13:00-13:50 in A 011 1 31 0160
- Exercises: Wednesday at 14:00 - 14:50 in A 011 1 31 0160
- Office Hours:
- Sundays,Tuesdays from 8am to 11am and Wednesdays from 8am to 12 pm.
- Email: Always available.
Prerequisites
- Official: CSC 361: Artificatial intelegence
- Unofficial: Basic probability and statistics, basic linear algebra
Course Description
CSC 462 is an introductory course to Machine Learning. The course will cover the following topics: Decision-tree learning; Ensemble learning; Statistical learning Methods: Bayes decision models, learning Bayes networks, Hidden Markov Models; Instance based learning; Neural Networks; Reinforcement learning; Clustering; Computational learning theory.
The objective of the course is between the theoretical and the practical spectrum. The concepts behind the above machine learning algorithms will be studied without going deeply into the mathematics behind them in order to gain more practical experience applying them. Both pattern recognition and artificial intelligence perspectives will be introduced in order to make the course attractive and helpful to all students interested in data science, engineering, and intelligent agent applications.
Textbook
-
Ethem Alpaydin, Introduction to Machine Learning, Second Edition http://mitpress.mit.edu/catalog/item/default.asp?ttype=2&tid=12012
Homework and Exams
Homework will be assigned every Wednesday to help you prepare for the exams. They will not be graded, but we will review the answers in class. There will be two Midterms and one Final exam.
Projects
In order to learn the practical aspect of machine leaning. We will have three projects. Each project is designed to highlight specific conceptual and practical issues that will help guide you to use machine learning for your own problems.
- Project1: Supervised Learning
- Project2: Unsupervised and Supervised Learning
- Project3: Hidden Markov Models and Reinforcement Learning
Grading
- Projects: 15% individual project, 25% group project.
- Midterm Exam: 20%
- Final Exam: 40%
Resources
Class resources will be maintained on this web site. Projects will be submitted and grades will be maintained on LMS.
- Weka - Machine learning software you'll be using for some of your projects.
- UCI Machine Learning Repository - An online repository of data sets that can be used for machine learning experiments.
Lectures and Assignments
Schedule subject to change.
| Week | Topics | Assignments |
| 1 | Introduction | Reading: Chapter 1 Assignment: 1.1,2,3 |
| 2 | Supervised Learning | Reading: Chapter 2 Assignment: 2.1,2,3,4,7 |
| 3 | Baysian Decision Theory | Reading: Chapter 3 Assignment: 3.3,9 |
| 4 | Nonparametric Methods | Reading: Chapter 8 Assignment: 8.3,4 |
| 5 | Decision Trees Machine Learning Experiments Weka Tutorial |
Reading: Chapter 9 Reading: Chapter 19 Assignment: Project 1 |
| Exam 1: November 10th, 2014 | ||
| 6 | Linear Discrimination Tutorial |
Reading: Chapter 10 Assignment: 10.1,7-9 |
| 7 | Multilayer Perceptrons Face Recognition using NN |
Reading: Chapter 11 Assigment: Project 2 |
| 8 | Dimmensionality Reduction | Reading Assigment |
| 9 | Clustering | Reading Assigment |
| 10 | Kernal Machines | Reading Assigment |
| 11 | Combining Learners | Reading Assigment |
| 12 | Reinforcement Learning | Reading Assigment: |
| Final Exam: January 6th, 2015. From 8:00 am to 11:00 am |