TY - JOUR
T1 - Enhancing supervised surface reconstruction through implicit weight regularization
AU - Noorizadegan, Amir
AU - Hon, Yiu Chung
AU - Young, Der Liang
AU - Chen, Chuin Shan
N1 - The authors gratefully acknowledge the financial support of the National Science and Technology Council of Taiwan under grant numbers 112-2221-E-002-097-MY3, 112-2811-E-002-020-MY3. We also want to acknowledge the NTUCE-NCREE Joint Artificial Intelligence Research Center and the National Center of High-performance Computing (NCHC) in Taiwan for providing computational and storage resources. The first author was supported by “National Group for Scientific Calculation” (GNCS - INDAM), Italy .
Publisher Copyright:
© 2025 Elsevier Ltd
PY - 2025/11
Y1 - 2025/11
N2 - Regularization plays a crucial role in stabilizing neural network training and improving generalization. In this work, we introduce the Square Highway (SqrHw) network, a novel variant of the Highway network that implicitly regularizes weight updates via a modified affine transformation. We analyze the effectiveness of the proposed regularization technique in supervised function approximation for surface construction tasks. These tasks involve large, structured datasets and clear evaluation metrics, which allow for a clearer observation of the method's performance. Our results show that SqrHw achieves superior convergence and improved stability in weight updates, outperforming both plain MLPs and cutting edge benchmark weight normalization techniques. Additionally, we conduct principal component analysis and gradient analysis to gain deeper insights into how the proposed network enhances robustness. The proposed architecture has broad applicability beyond supervised surface reconstruction, extending to fields where MLPs are commonly employed. The implementation is available at: https://github.com/AmirNoori68/SqrHw.git
AB - Regularization plays a crucial role in stabilizing neural network training and improving generalization. In this work, we introduce the Square Highway (SqrHw) network, a novel variant of the Highway network that implicitly regularizes weight updates via a modified affine transformation. We analyze the effectiveness of the proposed regularization technique in supervised function approximation for surface construction tasks. These tasks involve large, structured datasets and clear evaluation metrics, which allow for a clearer observation of the method's performance. Our results show that SqrHw achieves superior convergence and improved stability in weight updates, outperforming both plain MLPs and cutting edge benchmark weight normalization techniques. Additionally, we conduct principal component analysis and gradient analysis to gain deeper insights into how the proposed network enhances robustness. The proposed architecture has broad applicability beyond supervised surface reconstruction, extending to fields where MLPs are commonly employed. The implementation is available at: https://github.com/AmirNoori68/SqrHw.git
KW - Deep learning
KW - Highway network
KW - Principal component analysis
KW - Supervised surface reconstruction
KW - Weight regularization
UR - https://www.scopus.com/pages/publications/105015636338
UR - https://www.sciencedirect.com/science/article/pii/S0955799725003273?via%3Dihub
U2 - 10.1016/j.enganabound.2025.106439
DO - 10.1016/j.enganabound.2025.106439
M3 - Journal article
AN - SCOPUS:105015636338
SN - 0955-7997
VL - 180
JO - Engineering Analysis with Boundary Elements
JF - Engineering Analysis with Boundary Elements
M1 - 106439
ER -
