Rechnergestützte Intelligenz im Ingenieurwesen
Steckbrief
Eckdaten
- Abschluss:
- Master
- Semester:
- Wintersemester
- Dozent:
- Univ.-Prof. Dr.-Ing. Bernd Markert
- Sprache:
- Englisch
Weitere Informationen
Kontakt
Telefon
- work
- +49 241 80 94600
- E-Mail schreiben
Der Wahlpflichtkurs "Computational Intelligence in Engineering" ist für Studierende der ingenieurwissenschaftlichen Masterstudiengänge der RWTH Aachen verfügbar. Dieser gibt einen Überblick über aktuelle Anwendungen von Computational Intelligence und vertiefendes Lernen, die für die Ingenieurwissenschaften relevant sind. Der Kurs wird interaktiv unterrichtet, wobei die Studierenden an praktischen Beispielprojekten teilnehmen.
Themen
- Datenerfassung und -vorverarbeitung
- Feature-Skalierung
- Analyse der Hauptkomponenten
- Selbstorganisierende Karten und K-Mittel
- Analysieren von statischen Daten
- Lineare und logistische Regression
- Vektorielle Maschinen unterstützen
- Vollständig angeschlossene neuronale Netzwerke
- Analyse zeitvarianter Daten aus inertialen Messeinheiten (z. B. in Smartphones)
- Wiederkehrende neuronale Netze
- Analysieren von Raumdaten aus Simulationen
- Neuronale Faltungsnetze
Zusatzinformationen
Course Requirements
- Programming experience is requested (Python)
► Computer lab exercises are accompanied by extra lessons and tutorials on Python programming language
Course Objectives
The students will get an overview over current trends in computational intelligence (CI) and understand their theoretical foundation. They will be able to apply machine learning methods to a wide variety of engineering problems. The practical expertise gained by the students will enable them to transfer their knowledge to new engineering applications in science and industry. It is the ultimate objective of the course to qualify the students to evaluate the merits and limitations of CI methods in real engineering applications, such as predictive maintenance, structural health monitoring or modelling and simulation in Industry 4.0 projects among others.
Teaching and learning methods
The course curriculum consists of interactive seminar lectures accompanied by two semester projects. During the seminar lectures, the students will receive the necessary theoretical information and supervision to independently plan, advance and complete the projects in small groups. In addition, the seminars offer the opportunity to discuss challenges and problems arising during the course of the projects. Finally, the achievements and results obtained within the student projects will be presented by the students in the scope of the seminars and the accompanying computer lab exercises.
Recommended literature
Goodfellow, I., Bengio, Y., Courville, A., 2016. Deep Learning. MIT Press.
Schölkopf, B., Smola, A.J., Scholkopf, M.D. of the M.P.I. for B.C. in T.G.P.B., Bach, F., 2002. Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond. MIT Press.
Keller, J.M., Liu, D., Fogel, D.B. , 2016. Fundamentals of Computational Intelligence. IEEE Press, Wiley.