Laboratory of Computer and Information Science / Neural Networks Research Centre CIS Lab Helsinki University of Technology

Näitä sivuja ei päivitetä enää. Ole hyvä ja katso tietojenkäsittelytieteen laitoksen WWW-sivuja: http://ics.tkk.fi/fi/studies/.

These pages are not any more updated. Please, see web pages of Department of Information and Computer Science (ICS): http://ics.tkk.fi/en/studies/.

Näitä sivuja ei päivitetä enää. Ole hyvä ja katso tietojenkäsittelytieteen laitoksen WWW-sivuja: http://ics.tkk.fi/fi/studies/.

These pages are not any more updated. Please, see web pages of Department of Information and Computer Science (ICS): http://ics.tkk.fi/en/studies/.

T-61.5030 Advanced Course in Neural Computing (5 cp) P, Autumn 2005




KURSSI POISTUNUT OPETUSOHJELMASTA!!!

Kurssi luennoitiin viimeisen kerran keväällä 2007. Viimeinen tentti keväällä 2008. Korvaava kurssi T-61.5130 Machine Learning and Neural Networks


The course was lectured last time in Spring 2007. The last exam in Spring 2008. Replacing course T-61.5130 Machine Learning and Neural Networks
 T-61.5030 Neuraalilaskennan jatkokurssi

Contents and background

The course is a direct continuation of the basic course T-61.3030 Principles of Neural Computing. This advanced course proceeds at a faster pace than the basic course, dealing with fairly large number of matters. The topics include theory of neural learning, committee machines, principal component analysis, independent component analysis, stochastic machines based on statistical mechanics, neurodynamical programming, temporal processing, and recurrent networks. The basic course T-61.3030 or basic knowledge on neural networks is recommended as a prerequisite for this course. Furthermore, the participants should have a knowledge of basic university mathematics, including fundamentals of calculus, linear algebra, and probability theory.

Material

The textbook is S. Haykin, Neural Networks - A Comprehensive Foundation, 2nd ed., Prentice-Hall, 1998. The first chapters of the book on fundamental methods of neural computing are discussed in the basic course. Exercise problem sheets will be available on the Web pages of the course, as well as their solutions after the exercises. The exercises, their solutions, examination requirements, lecture slides written in English, as well as a review article on independent component analysis which is used as teaching material will be copied to the participants as lecture notes via Edita Prima Oy.

Lectures and exercises

The lectures will be given on Mondays 14-16 o'clock in the classroom T3 on the Dept. of Computer Science and Engineering. Solutions of the exercises will be presented on Thursdays 10-12 o'clock in the same classroom T3. Notice: exceptionally the first lecture will be held on Thursday 14th September 10-12 in the classroom T3. The first exercises will be on 21st September 2006. The lectures and exercises extend over the teaching periods I (Sept. 8 - Oct. 25) and II (Nov. 2 - Dec. 13) in autumn 2006. During the examination week (Oct. 26 - Nov. 1) between the teaching periods I and II there are no lectures and exercises. The last lecture on 11th Dec. is planned to be a demo lecture on selected applications related to the matters discussed in the course. Both the lectures and exercises will be presented orally in Finnish, but all the course materials are in English. This will make a self-study of the course possible also for non-Finnish participants.

The lectures will be given by Prof. Juha Karhunen. He can be met during the lectures, or by email: Juha.Karhunen@hut.fi (Tel. 09-451 3270, Computer Science House, Room B327). The exercises will be given by Dr. Jaakko Peltonen; email jaakko.peltonen@hut.fi, Tel. 09-451 4429, Computer Sci. House Room A326.

Examinations

The first examination will be arranged on Monday 18th December 16-19 o'clock, in the lecture hall T1. The second exam will tentatively be held on Sat. 17th February 2007, and the third one in Autumn 2007. If you need the exam paper in English please please notify the lecturer at least one week beforehand! Please enroll to the examination via wwwtopi a week before the examination at the latest. This course does not include computer assignments.

Notices and enrollment

The notice board of the course is in the 3rd floor of the T-building. Additionally, notices are given on the lectures and exercises, in the newsgroup opinnot.tik.informaatiotekniikka, and on the Web page of the course, http://www.cis.hut.fi/Opinnot/T-61.5030. Please enroll to the course using wwwtopi, or if this is not possible by e-mail to the lecturer.


Otaniemi, August 31, 2006

Prof. Juha Karhunen

News

  • The time and date of the first lecture have changed: the first lecture will be held on Thursday 14th September 10-12 in the classroom T3.
  • Slides of the first lecture in PDF format
  • A tutorial article on independent component analysis
  • The exercise session on Thursday, Nov. 30, 2006 (from 10am to 12am) will handle both the 10th and 11th exercise set. The week after that there will not be an exercise session any more.
  • On Monday Dec. 11, there will be a demo lecture from 14:15 to 15:30 in lecture hall T3 (in the Computer Science Building). There will be three talks:
    1. About 14:15-14:40: Matti Aksela, "Luokittelijoiden yhdistäminen käsinkirjoitettujen merkkien tunnistamiseksi". Slides (in english) in PDF format
    2. About 14:40-15:05: Alexander Ilin, "Finding interesting climate phenomena using source separation techniques". Slides in PowerPoint format (12.7 Mb)
    3. About 15:05-15:30: Tapani Raiko, "Variational Bayesian approach for nonlinear identification and control". Slides in PowerPoint format (2.1 Mb)
    The talks are in Finnish except for A. Ilin's talk.
  • Examination requirements in PDF format
  • Slides for the demo lecture are now available, see the hyperlinks above.

Exercises

You are at: CIS → T-61.5030 Advanced Course in Neural Computing (5 cp) P

Page maintained by t615030@cis.hut.fi, last updated Tuesday, 19-Aug-2008 10:51:03 EEST