Artificial neural networks can be most adequately characterized as computational models with particular properties such as the ability to adapt or learn, to generalize, or to cluster or organize data, and which operation is based on parallel processing. How-ever, many of the abovementioned properties can be also attributed to other (non neural) models. The intriguing question is to which extent the neural approach proves to be better suited for certain applications than existing models. Thus, NNs can be generally considered as an alternative computational scheme.

Relevant publications:

Plevris, Vagelis; Ramirez, German Solorzano; Bakas, Nikos. Literature review of historical masonry structures with machine learning. I: Proceedings of the 7th International Conference on Computational Methods in Structural Dynamics and Earthquake Engineering (COMPDYN 2019). European Community on Computional Methods in Applied Sciences (ECCOMAS) 2019 ISBN 978-618-82844-5-6. s. 1547-1562

Back to Research