TY - GEN
T1 - Exploring Various Sequential Learning Methods for Deformation History Modeling
AU - Yatkin, Muhammed Adil
AU - Korgesaar, Mihkel
AU - Romanoff, Jani
AU - Stuckner, Joshua
AU - Işlak, Ümit
AU - Kurban, Hasan
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.
PY - 2025/6/22
Y1 - 2025/6/22
N2 - Current neural network (NN) models can learn patterns from data points with historical dependence. Specifically, in natural language processing (NLP), sequential learning has transitioned from recurrence-based architectures to transformer-based architectures. However, it is unknown which NN architectures will perform the best on datasets containing deformation history due to mechanical loading. Thus, this study ascertains the appropriateness of 1D-convolutional, recurrent, and transformer-based architectures for predicting deformation localization based on the earlier states in the form of deformation history. Following this investigation, the crucial incompatibility issues between the mathematical computation of the prediction process in the best-performing NN architectures and the actual values derived from the natural physical properties of the deformation paths are examined in detail.
AB - Current neural network (NN) models can learn patterns from data points with historical dependence. Specifically, in natural language processing (NLP), sequential learning has transitioned from recurrence-based architectures to transformer-based architectures. However, it is unknown which NN architectures will perform the best on datasets containing deformation history due to mechanical loading. Thus, this study ascertains the appropriateness of 1D-convolutional, recurrent, and transformer-based architectures for predicting deformation localization based on the earlier states in the form of deformation history. Following this investigation, the crucial incompatibility issues between the mathematical computation of the prediction process in the best-performing NN architectures and the actual values derived from the natural physical properties of the deformation paths are examined in detail.
KW - Localization in Sheet Metal
KW - Recurrent Neural Networks
KW - Sequential Learning
KW - Surrogate Modelling
UR - https://www.scopus.com/pages/publications/105009834331
U2 - 10.1007/978-3-031-96196-0_13
DO - 10.1007/978-3-031-96196-0_13
M3 - Conference contribution
AN - SCOPUS:105009834331
SN - 9783031961953
VL - 2581
T3 - Communications In Computer And Information Science
SP - 168
EP - 180
BT - Engineering Applications Of Neural Networks, Eann 2025, Pt I
A2 - Iliadis, L
A2 - Maglogiannis, I
A2 - Kyriacou, E
A2 - Jayne, C
PB - Springer Science and Business Media Deutschland GmbH
T2 - 26th International Conference on Engineering Applications of Neural Networks, EANN 2025
Y2 - 26 June 2025 through 29 June 2025
ER -