Title: Pattern Recognition
Catalog Description: Overview of learning and statistical decision theory. Model inference and parameter estimation. Linear models for regression and classification. Kernel methods. Nonparametric methods. Model assessment and selection. Ensemble methods. Unsupervised learning.
Coordinator: H. Işıl Bozma, Prof.
Goals: This course is designed for seniors and graduate students. The students will be exposed to contemporary pattern recognition and machine learning approaches. The emphasis of the course is not only the theory but also practical applications.
At the end of this course the students will be able to:
· Build pattern recognition systems
· Compare various methods for classification and regression
· Understand the basics of machine learning
Textbook: Duda, Hart, Stork, Pattern Classification, 2nd Ed., Wiley-Interscience, 2004.
· Hastie, Tibshirani, Friedman, The Elements of Statistical Learning, Springer, 2001.
· Bishop, Pattern Recognition and Machine Learning, 2006.
· Ethem Alpaydin, Introduction to Machine Learning, 2004.
Prerequisite by Topic:
Overview of Pattern Recognition
Bayesian Decision Theory
Parameter Estimation: ML, Bayesian
Linear Regression: Least Squares, Subset Selection and Shrinkage
Linear Discriminant Methods for Classification: LDA, Perceptron, SVM
Dimensionality Reduction: PCA, Reduced Rank LDA
Kernel Methods: Support Vector Machines and Support Vector Regression
Nonparametric Methods: Parzen Windows, Nearest Neighbors
Model Assessment and Selection: BIC, MDL, VC-Dimension, Cross Validation
Unsupervised Learning: Clustering, Mixture Models
The class meets for two lectures a week -- one lecture consists of a two-hour session and the second lecture is one-hour. The students are required to complete five projects as part of requirements.In the projects, students are expected to first prepare a literature survey on the topic and then to apply the techniques that they learned in the class.
Projects require MATLAB programming
Final Examination %30
a. an ability to apply knowledge of mathematics and computer programming: knowledge of mathematics is used to describe and analyze various machine learning algorithms where as knowledge of computer programming is used to implement the algorithms.
b. an ability to design and conduct experiments, as well as to analyze and interpret data: proper use of data for conducting experiments is emphasized, various methods for unsupervised data analysis methods are studied.
c. an ability to design a pattern recognition system.
Prepared By: H. Işıl Bozma