courses:ae4b33rpz:lectures:start [Course Ware]

Annotation

The basic formulations of the statistical decision problem are presented. The necessary knowledge about the (statistical) relationship between observations and classes of objects is acquired by learning on the training set. The course covers both well-established and advanced classifier learning methods such as Perceptron, AdaBoost, Support Vector Machines, and Neural Nets.

Basic info

Winter semester 2011/2012

Where and when: KN:G-205 at Building G on Charles square, Monday 14:30-16:00

Teaching: Prof. Jiří Matas matas@cmp.felk.cvut.cz

Lectures plan

Week	Date	Lecturer	Topic
1	19.9.	J. Matas	Introduction. Recognition tasks formulation. Map of the subject. Basic notions. (pdf)
2	26.9.	V. Franc	Bayesian decision task (pdf)
3	3.10.	J. Matas	Non-Bayesian tasks (pdf)
4	10.10.	V. Franc	Parameter estimation of probabilistic models. Maximum likelihood method. (slides by JM) (wiki)
5	17.10.	J. Matas	Nearest neighbour method. Non-parametric density estimation. (ppt)
6	24.10.	J. Matas	Classifier training. Linear classifier. Perceptron. (pdf)
7	31.10.	J. Matas	Classifier training as a quadratic optimisation task. SVM classifiers. (ppt) (pdf) (pdf by VH)
8	7.11.	J. Matas	Adaboost learning (pdf)
9	14.11.	J. Matas	Neural networks. Backpropagation (pdf)
10	21.11.	J. Matas	Cluster analysis, k-means method (pdf)
11	28.11.	J. Matas	Unsupervised learning. EM (Expectation Maximization) algorithm. (pdf by Hoffman) (ppt Bishop)
12	5.12.	J. Matas	Feature selection and extraction. PCA, LDA.(pdf)
13	12.12.	J. Matas	Basic notions recapitulation, links between methods, answers to exam questions

Recommended literature

Duda R.O., Hart, P.E.,Stork, D.G.: Pattern Classification, John Willey and Sons, 2nd edition, New York, 2001
Schlesinger M.I., Hlaváč V.: Ten Lectures on Statistical and Structural Pattern Recognition, Springer, 2002

Exam

Only those who got credit (“zápočet”) from the labs can participate in the exam. The labs contribute to the evaluation by at most 50%.
The exam consists of two parts: written test and oral exam.
Written test lasts 60 minutes and contributes to the evaluation by 40%.
The questions used in the test are available here (if one can solve these questions, one will very likely succeed during the exam)
Oral part starts approximately 2 hours after the end of the test (time is used to correct the tests). It contributes to the final evaluation by 10%.
Oral exam questions are available here.