Modul Artificial Neural Networks

Code:	214350
Module title:	Artificial Neural Networks
Version:	1.0 (03/2016)
Last update:	13.02.2024
Responsible person:	Prof. Dr.-Ing. Kästner, Wolfgang w.kaestner@hszg.de

Offered in 4 study courses:	Automation and Mechatronics (B.Eng.) valid from class 2018
	Automation and Mechatronics (B.Eng.) valid from class 2021
	Electrical Engineering (B.Eng.) valid from class 2024
	Mechatronics (M.Eng.) valid from class 2019

Semester according to time table:	WiSe (winter semester)
Module level:	Bachelor/Diplom
Duration:	1 semester
Status:	elective core module (specialization)
Place where the module will be offered:	Zittau
Language of Instruction:	English

Workload* in

SCH **

semester

hours

ECTS Credits

1

2

3

4

5

6

7

L

S

P

O

L

S

P

O

L

S

P

O

L

S

P

O

L

S

P

O

L

S

P

O

L

S

P

O

150

5

4.0

2

1

1

0

*	Overall workload per module (1 ECTS credit corresponds to a workload of 30 hours)
**	One semester credit hour (SCH) corresponds to a workload / class meeting of 45 minutes per week in a semester

Self study time in hours	total	subdivided into
	105	70 Preperation of contact hours	35 Preparation of exam	0 Others

Learning and teaching methods:	The methodical aspects of the topic will be communicated by lectures. Seminars and exercises as well as practical courses at laboratory (PC tool) serve for consolidation of knowledge.

Further information:

PC-based exercises will be realized to train the handling of simulation tools.



Exam(s)
Assessment	Major examination (written report)	-	100.0%

---

Syllabus plan/Content:	foundations of Artificial Neural Networks (ANN) application of ANN for modelling and classification modelling using Multilayer Perceptron (MLP) MLP - structure, demonstration example, software classification using on Kohonen Maps (SOM) SOM - structure, demonstration example, software applications simulation tools

Learning Outcomes:
Subject-specific skills and competences:	The students analyze a data base of process and identify the necessity of data preparation. They design data-based models in form of multilayer perceptron as well as classifiers in form of self-organizing maps. They evaluate the quality of designed algorithms based on simulation
Generic competences (Personal and key skills):	The students are able to create and realize strategies for problem solving from the individual point of view or as a result of teamwork. The students use approaches of system theory. The students evaluate their results and are able to present the results.

Pre-requisites:	competence from following module (without burden of proof): - Mathematics
Optional pre-requisites:	competence from module: - Signals and Systems

Literature:

Kruse, R. / Mostaghim, S. / Borgelt, C.: Computational Intelligence. Springer, 2022 Keller, J.: Computational Intelligence. John Wiley & Sons, 2016 Kroll, A.: Computational Intelligence. De Gruyter, 2016 Kruse, R. / Borgelt, C. / Braune, C. / Klawonn, F.: Computational Intelligence. Springer, 2015 Sonnet, D. Neuronale Netze kompakt. Springer, 2022 Ertel, W.: Grundkurs Künstliche Intelligenz. Springer, 2021 Lämmel, U. / Cleve, J.: Künstliche Intelligenz. Carl Hanser, 2020 Beierle, C. / Kern-Isberner, G.: Methoden wissensbasierter Systeme. Springer, 2019