Course Catalog 2010-2011
International

Basic Pori International Postgraduate Open University

|Degrees|     |Study blocks|     |Courses|    

Course Catalog 2010-2011

SGN-2806 Neural Computation, 5 cr

Person responsible

Ari Visa

Lessons

Study type P1 P2 P3 P4 Summer Implementations Lecture times and places
Lectures

 
 4 h/per

 

 

 
SGN-2806 2010-01 Tuesday 12 - 14, TB224
Thursday 14 - 16, TB222
Monday 12 - 14, TB224
Friday 10 - 16, TB224

Requirements

Final exam, attendance at the classroom exercises and assignment.
Completion parts must belong to the same implementation

Principles and baselines related to teaching and learning

-

Learning outcomes

Learning outcomes: The student can list the mentioned learning rules. The student can describe them and is capable to apply them to train neural networks. The student is capable to list to analyse the lectured neural networks (MLP,SVM,SOM and recurrent networks). The student is capable to analyse the own problem and to select the most suitable, lectured neural network. The student has a certain capability to create new solutions based on the lectured material.

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   

Evaluation criteria for the course

Exam and approx. 70% attendance at the exercises.

Assessment scale:

Numerical evaluation scale (1-5) will be used on the course

Partial passing:

Completion parts must belong to the same implementation

Study material

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

Prerequisite relations (Requires logging in to POP)



Correspondence of content

Course Corresponds course  Description 
SGN-2806 Neural Computation, 5 cr 8001703 Neural Computation, 3 cu  

Additional information

Lectures in English or in Finnish.
Suitable for postgraduate studies

More precise information per implementation

Implementation Description Methods of instruction Implementation
SGN-2806 2010-01        

Last modified11.02.2010