A Hybrid Bayesian-Machine Learning Approach for Inverse Cubic Sensor Calibration
i-manager’s Journal on Mathematics · 20.06.2026 · 15 (1) · 1
DOI: 10.26634/jmat.15.1.652 ↗Kurz gefasst
Fachartikel in der Zeitschrift i-manager’s Journal on Mathematics, online erschienen am 15.06.2026. Verfasst von Ganti Srikanth. Die Arbeit nennt 27 Literaturangaben.
Abstract (Originalsprache)
Cubic polynomial transformations are commonly employed in sensor systems to represent nonlinear responses, achieving a balance between precision and computational efficiency. The inverse problem-retrieving physical input from measured output is essential for sensor calibration but is complicated by multiple roots, noise sensitivity and parameter uncertainty. The work introduces a hybrid methodology that inteprates Bayesian inference with ensemble learning for cubic inverse problems. A theoretical justification for Machine Learning (ML)-based root selection using discriminant analyses and derivative constraints is brought out in the work. The results obtained are compared against analytical and numerical techniques on synthetic data sets with 80-85% multiple root cases. The proposed methodology achieves 90% root selection accuracy while reducing the reconstruction error by 34% compared to analytical methods. The noise sensitivity, uncertainty quantifications are also analyzed.
Zitieren
APA
Srikanth, G. (2026). A Hybrid Bayesian-Machine Learning Approach for Inverse Cubic Sensor Calibration. i-manager’s Journal on Mathematics, 15(1), 1. https://doi.org/10.26634/jmat.15.1.652
Vancouver
Srikanth G. A Hybrid Bayesian-Machine Learning Approach for Inverse Cubic Sensor Calibration. i-manager’s Journal on Mathematics. 2026;15(1):1. doi:10.26634/jmat.15.1.652
BibTeX
@article{Srikanth2026,
author = {Srikanth, Ganti},
title = {A Hybrid Bayesian-Machine Learning Approach for Inverse Cubic Sensor Calibration},
journal = {i-manager’s Journal on Mathematics},
year = {2026},
volume = {15},
number = {1},
pages = {1},
doi = {10.26634/jmat.15.1.652}
}
EUVIMED