Exponential Function-Based Neural Tangent Kernels for SECM Signal Reconstruction
Applied Sciences · 06.04.2026 · 16 (7) · 3578
DOI: 10.3390/app16073578 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 06.04.2026. Verfasst von Vadimas Ivinskij und 6 weiteren Autor:innen. Die Arbeit nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: EU.
Abstract (Originalsprache)
Scanning electrochemical microscopy (SECM) provides spatially resolved electrochemical information but is constrained by long acquisition times arising from dense spatial sampling requirements. This work investigates whether physics-informed signal representations can improve neural reconstruction of SECM approach curve signals from sparse measurements. We propose an exponential function-based Neural Tangent Kernel (NTK) framework in which SECM signals are encoded using deterministic exponential feature mappings aligned with diffusion-controlled electrochemical dynamics. A layer-wise NTK checkpointing mechanism is employed to filter covariantly insignificant components during training, reducing redundancy while preserving dominant signal modes. The method is evaluated on synthetically generated SECM signals designed to replicate characteristic approach curve behavior. Quantitative performance is assessed using root mean square error (RMSE), mean absolute error (MAE), relative error (%), and the coefficient of determination (R2). Compared to a random Gaussian (Fourier feature) baseline (RMSE = 0.0952, MAE = 0.0547, Rel.Err = 17.68%), the proposed exponential mappings achieve consistently lower reconstruction error, with the best configuration yielding RMSE = 0.0858, MAE = 0.0375, and relative error = 11.10% under identical training conditions. Results demonstrate that incorporating physically motivated exponential feature representations into NTK-aware learning improves reconstruction fidelity and stability for low-dimensional electrochemical signals, highlighting the potential of physics-informed kernel methods for accelerated SECM data acquisition.
Förderung
- EU
Zitieren
APA
Ivinskij, V., Mačerauskas, E., Striška, L., Plonis, D., Pandey, V., Tolvaisiene, S., & Morkvėnaitė, I. (2026). Exponential Function-Based Neural Tangent Kernels for SECM Signal Reconstruction. Applied Sciences, 16(7), 3578. https://doi.org/10.3390/app16073578
Vancouver
Ivinskij V, Mačerauskas E, Striška L, Plonis D, Pandey V, Tolvaisiene S, et al. Exponential Function-Based Neural Tangent Kernels for SECM Signal Reconstruction. Applied Sciences. 2026;16(7):3578. doi:10.3390/app16073578
BibTeX
@article{Ivinskij2026,
author = {Ivinskij, Vadimas and Mačerauskas, Eugenijus and Striška, Laisvidas and Plonis, Darius and Pandey, Vijitashwa and Tolvaisiene, Sonata and Morkvėnaitė, Inga},
title = {Exponential Function-Based Neural Tangent Kernels for SECM Signal Reconstruction},
journal = {Applied Sciences},
year = {2026},
volume = {16},
number = {7},
pages = {3578},
doi = {10.3390/app16073578}
}
EUVIMED