Course Catalog 2006-2007

SGN-2806 NEURAL COMPUTATION, 5 cr
Neural Computation

Courses persons responsible
Ari Visa

Lecturers
Ari Visa

Lecturetimes and places
Per III: Monday 10 - 12, TB222
Per III: Wednesday 14 - 16, TB223

Implementations
  Period 1 Period 2 Period 3 Period 4 Period 5 Summer
Lecture - - 4 h/week - - -
Exercise - - 2 h/week - - -
Assignment - - 20 h/per - - -
Exam  
(Timetable for academic year 2006-2007)

Objectives
To give basic knowledge of neuro computing and to apply neuro computing to some application fields.

Content
Content Core content Complementary knowledge Specialist knowledge
1. Learning processes       
2. Learning machines with a teacher  Multilayer Perceptrons
Radial-Basis Function Networks
Support Vector Machines
Committee Machines 		   
3. Learning machines without a teacher  Principal Component Analysis with Neural Networks
Self-Organizig Maps
Boltzmann Machine 		   
4. Nonlinear dynamical systems  Temporal Processing Using Feed Forward Network
Dynamically Driven Recurrent Network 		   

Requirements for completing the course
Final exam, attendance at the classroom exercises and assignment.

Evaluation criteria for the course

  • Exam and approx. 70% attendance at the exercises.

  • Used assessment scale is numeric (1-5)

    • Study material
    Type Name Auhor ISBN URL Edition, availability... Exam material Language
    Book "Neural Networks: a Comprehensive Foundation" Haykin, S.     2nd edition, Prentice-Hall Inc, 1999 Yes  English 

    Prerequisites
    Code Course Credits M/R
    MAT-31090 MAT-31090 Matrix Algebra 1 5 Recommendable
    MAT-41120 MAT-41120 Optimization Theory 1 5 Recommendable

    Prequisite relations (Sign up to TUT Intranet required)

    Remarks

    The course is intended to students who are close to graduation in the fields of signal processing, computer science or telecommunication. Lectures in English or in Finnish.

  • Partial passing of course must be in connection with the same round of implementation.

  • The course is suitable for postgraduate studies.

    • Distance learning

  • ITC utilized during the course

    • - In information distribution via homepage, newsgroups or mailing lists, e.g. current issues, timetables
    - In distributing and/or returning exercise work, material etc
    - In the visualization of objects and phenomena, e.g. animations, demonstrations, simulations, video clips

  • Estimate as a percentage of the implementation of the course

    • - Contact teaching: 38 %
    - Distance learning: 0 %
    - Proportion of a student's independent study: 62 %

    Scaling
    Methods of instructionHours
    Lectures 72
    Exercises 28

    Other scaledHours
    Preparation for exam 30
    Exam/midterm exam 3
    Total sum 133

    Principles and starting points related to the instruction and learning of the course

  • Students are encouraged to ask questions both during/after the lectures and exercises.

    • Additional information related to course
    Lectures in English or in Finnish.

    Correspondence of content
    8001703 Neural Computation

    Course homepage

    Last modified 02.02.2006
    Modified byAntti Niemistö