Login
English
Nederlands
Neural Networks and Computational Intelligence
| Faculteit | Science and Engineering |
| Jaar | 2022/23 |
| Vakcode | WMCS010-05 |
| Vaknaam | Neural Networks and Computational Intelligence |
| Niveau(s) | master |
| Voertaal | Engels |
| Periode | semester I b |
| ECTS | 5 |
| Rooster | rooster.rug.nl |
| Uitgebreide vaknaam | Neural Networks and Computational Intelligence | ||||||||||||||||
| Leerdoelen | At the end of the course, the student is able to: 1) implement and use various types of artificial neural network 2) relate artificial neural networks to their biological background and different levels of modelling biological systems 3) explain, express mathematically, implement and apply basic training schemes 4) explain the concept of supervised learning from examples and are able to apply it in terms of simple example situations 5) to explain and to implement and make use of standard validation and regularization methods in practical situations 6) perform computer simulations of machine learning processes and present and discuss results thereof according to scientific standards. 7) work in teams (pairs) on scientific projects and reports |
||||||||||||||||
| Omschrijving | This module provides an introduction to neural networks and related concepts in machine learning. We will discuss different types of network architectures and their usefulness and limitations in classification or regression problems. In this context, the corresponding training algorithms will be the focus of our attention. Besides their practical implementation, we will address theoretical aspects, e.g. with respect to their convergence behaviour. The list of topics includes: perceptron training, support vector machines, gradientbased training, testing and validation methods, multilayered neural networks, shallow and deep networks, alternative architectures. | ||||||||||||||||
| Uren per week | |||||||||||||||||
| Onderwijsvorm | Hoorcollege (LC), Practisch werk (PRC) | ||||||||||||||||
| Toetsvorm |
Schriftelijk tentamen (WE), Verslag (R)
(Final grade (FG): weighted average (WA) of the grade for the exam (50%) and an averaged grade for three assignments (50%). WAs above 4.75 and below 5.75 will result in a FG of 5.0, in all other cases the FG is rounded to the nearest half-integer. However, in order to pass the course, both partial grades have to be at least 5.5. If this is not the case, the FG is 5.0 or lower according to the WA.) |
||||||||||||||||
| Vaksoort | master | ||||||||||||||||
| Coördinator | Prof. Dr. M. Biehl | ||||||||||||||||
| Docent(en) | Prof. Dr. M. Biehl , M. Mohammadi | ||||||||||||||||
| Verplichte literatuur |
|
||||||||||||||||
| Entreevoorwaarden | The course unit assumes prior knowledge in the sense that basic programming skills in one of the major programming languages and/or tools like Matlab or Mathematica have been acquired. Having attended courses like Introduction to Intelligent Systems or Introduction to Artificial Intelligence is benefitial but not required. | ||||||||||||||||
| Opmerkingen | In order to pass the course, both partial grades (exam and averaged practicals) have to be at least 5.5, individually. If this is not the case, the final grade is equal to 5.0 or lower according to the lowest of the two partial grades. The "written exam" is realized as a digital exam. This course has limited enrollment: - CS students can always enter the course, regardless of whether the course is mandatory for them or not. - The number of enrolments for other non-CS students is limited. These students need to meet the course prerequisite requirements as mentioned on Ocasys. Priority is given to students for which the course is an official elective (see list below). - An exception can be made for exchange students, if they have a CS background: please contact the FSE International Office. See here for more info about the enrollment procedure. |
||||||||||||||||
| Opgenomen in |