Nima Hatami, Ph.D.
Applied machine learning, pattern recognition and computer vision (Spring 2014 & Fall 2015)
This introductory course will give an overview of many concepts, techniques, and algorithms in machine learning, pattern recognition and computer vision with many hands on coding experiments. The course will give the student the basic ideas and intuition behind modern machine learning methods as well as a bit more formal understanding of how, why, and when they work. At the end of this course, you are expected to be able to use state-of-the-art algorithms on wide range of applications.
Lecture. Thursdays 1-4pm, 3rd foor, EECS building.
Level. Masters in informatics, electronics, computer science and engineering
Requirements. Basic knowledge of programming, linear algebra, statistics and probability.
Codings. About %30 of each class will be devoted to programming. The students are encouraged to use any language they are more confident with e.g. Matlab, C/C++, python, R. So, bring your laptop with you and let's have some fun!
Lesson 1. Welcome! slides
Lesson 2. Introduction to pattern recognition and machine learning. slides
Lesson 3. Supervised learning. slides
Lesson 3.1 Principal Component Analysis and Linear Discriminant Analysis. slides
Lesson 4. Introduction to Multi-Layer Perceptron. slides
Lesson 4.1 Support Vector Machines and kernel learning. slides
Lesson 4.2 Multiple Classifier Systems
Lesson 5. Image formation and filtering. slides
Lesson 5.1 Stereo vision. slides
Lesson 6. Local image features: detection and description. slides
Lesson 7. Bag of words for visual recognition. slides
Lesson 8. Learning sparse representations. slides
Lesson 9. Learning feature hierarchies and deep learning. exercise:step-by-step deep learning with python