TMPNN: High-Order Polynomial Regression Based on Taylor Map Factorization
DOI:
https://doi.org/10.1609/aaai.v38i11.29168Keywords:
ML: Classification and Regression, ML: Transparent, Interpretable, Explainable MLAbstract
The paper presents Taylor Map Polynomial Neural Network (TMPNN), a novel form of very high-order polynomial regression, in which the same coefficients for a lower-to-moderate-order polynomial regression are iteratively reapplied so as to achieve a higher-order model without the number of coefficients to be fit exploding in the usual curse-of-dimensionality way. This method naturally implements multi-target regression and can capture internal relationships between targets. We also introduce an approach for model interpretation in the form of systems of differential equations. By benchmarking on Feynman regression, UCI, Friedman-1, and real-life industrial datasets, we demonstrate that the proposed method performs comparably to the state-of-the-art regression methods and outperforms them on specific tasks.Downloads
Published
2024-03-24
How to Cite
Ivanov, A., & Ailuro, S. (2024). TMPNN: High-Order Polynomial Regression Based on Taylor Map Factorization. Proceedings of the AAAI Conference on Artificial Intelligence, 38(11), 12726-12734. https://doi.org/10.1609/aaai.v38i11.29168
Issue
Section
AAAI Technical Track on Machine Learning II