COMPUTATIONAL COMPLEXITY OF MACHINE LEARNING ALGORITHMS: A MATHEMATICAL PERSPECTIVE
Scientific works/Elmi eserler · 22.07.2026 · 574
DOI: 10.58225/mpmma.2026.574 ↗Kurz gefasst
Fachartikel in der Zeitschrift Scientific works/Elmi eserler, online erschienen am 22.06.2026. Verfasst von Murad İsmayılov. Die Arbeit nennt 24 Literaturangaben.
Abstract (Originalsprache)
This paper presents a mathematical analysis of the computational complexity of fundamental machine learning algorithms. We examine time and space complexity of linear regression, support vector machines, decision trees, and neural networks, using asymptotic notation as the primary analytical tool. The empirical risk minimization framework and gradient-based optimization are discussed as unifying mathematical foundations. The study demonstrates that algorithm selection must be guided not only by predictive accuracy but also by complexity constraints imposed by the scale of the data. Our analysis draws on established theoretical frameworks and recent empirical comparisons to provide practical guidelines for algorithm deployment in data-intensive settings
Zitieren
APA
İsmayılov, M. (2026). COMPUTATIONAL COMPLEXITY OF MACHINE LEARNING ALGORITHMS: A MATHEMATICAL PERSPECTIVE. Scientific works/Elmi eserler, 574. https://doi.org/10.58225/mpmma.2026.574
Vancouver
İsmayılov M. COMPUTATIONAL COMPLEXITY OF MACHINE LEARNING ALGORITHMS: A MATHEMATICAL PERSPECTIVE. Scientific works/Elmi eserler. 2026:574. doi:10.58225/mpmma.2026.574
BibTeX
@article{smaylov2026,
author = {İsmayılov, Murad},
title = {COMPUTATIONAL COMPLEXITY OF MACHINE LEARNING ALGORITHMS: A MATHEMATICAL PERSPECTIVE},
journal = {Scientific works/Elmi eserler},
year = {2026},
pages = {574},
doi = {10.58225/mpmma.2026.574}
}
EUVIMED