BEGIN:VCALENDAR
METHOD:PUBLISH
PRODID:www-ww-tf-fau-eu//Events//EN
VERSION:2.0
BEGIN:VEVENT
SUMMARY:WW-Colloquium: Prof. Dr. Norbert Huber - Machine learning in m
 aterials science and engineering - best practice\, perspectives and pi
 tfalls
UID:040000008200E00074C5B7101A82E00800000000A0A602D6A683DC010000000000
 00000010000000EAE7ABDF41363D43B27E9D1F237ADD3C
DESCRIPTION:Prof. Dr. Norbert Huber Bundesanstalt für Materialforschu
 ng und -prüfung (BAM)\, Berlin and Institute of Materials Physics and
  Technology\, Hamburg University of Technology Machine learning in mat
 erials science and engineering &#8211\; best practice\, perspectives a
 nd pitfalls Machine learning (ML) is increasingly utilized to support 
 the data driven analysis of relationships in multidimensional paramete
 r spaces\, ideally as an entry point for a more general phenomenologic
 al or physics-based model development. Applications include both forwa
 rd and inverse problems as well as forward problems\, for example para
 meter identification or modeling of structure-property relationships. 
 The talk will give an overview over a variety of solutions that benefi
 t from the capability of artificial neural networks to approximate and
  interpolate complex relationships that are represented by a set of sp
 arse data. The reason behind is that numerical simulations as well as 
 experiments do often not allow to generate enough data such that the d
 ata set is not sufficient for a deep-learning approach in connection w
 ith the complexity of the problem at hand. After a short introduction 
 to artificial neural networks along with recommendations for data gene
 ration and feature engineering\, the talk will cover a range of exampl
 es from nanoindentation and material parameter identification\, the im
 provement of characterization techniques by ML correction methods towa
 rds recent problems in the prediction of structure-property relationsh
 ips for materials with complex microstructure. All these examples have
  in common that a successful ML model typically requires a comprehensi
 ve understanding of existing knowledge\, expertise in translating this
  knowledge into meaningful input features\, a compact ML architecture\
 , and robust validation of the trained model. The talk will conclude w
 ith the example of nanoporous metals that demonstrates the importance 
 of high-quality and bias-free data for the applicabili
DTSTART:20260113T150000Z
DTEND:20260113T170000Z
LOCATION:H14 / Zoom
DTSTAMP:20260414T112143Z
END:VEVENT
END:VCALENDAR