Work-in-Progress: Evaluating Feasibility of Band Matrix Solvers for Scaling up Extreme Learning Machine Method

Anton Akusok; Kaj-Mikael Björk; Amaury Lendasse; Leonardo Espinosa Leal

doi:10.1007/978-3-032-07319-8_31

Work-in-Progress: Evaluating Feasibility of Band Matrix Solvers for Scaling up Extreme Learning Machine Method

Anton Akusok^*
, Kaj-Mikael Björk
, Amaury Lendasse
, Leonardo Espinosa Leal

^*Corresponding author for this work

Not published at VTT

Arcada University of Applied Sciences

Research output: Chapter in Book/Report/Conference proceeding › Chapter or book article › Scientific › peer-review

Abstract

This work presents the results of the potential of band linear system solvers for improving the scalability of the Extreme Learning Machine (ELM) method at large model sizes. The model is tested on the standard MNIST dataset with a range of solvers provided by the SciPy Python library. The results are analyzed taking into consideration the overall performance and the performance impact of band solvers across different matrix bandwidths, as well as the performance versus runtime analysis. The findings show potential in applying the proposed method to very large ELM models with narrow band matrices.

Original language	English
Title of host publication	Smart Technologies for an All-Electric Society
Subtitle of host publication	Proceedings of the 22nd International Conference on Smart Technologies & Education (STE2025)
Publisher	Springer
Pages	347-355
Volume	2
ISBN (Electronic)	9783032073198
ISBN (Print)	9783032073181
DOIs	https://doi.org/10.1007/978-3-032-07319-8_31
Publication status	Published - 2026
MoE publication type	A3 Part of a book or another research book

Access to Document

10.1007/978-3-032-07319-8_31

Cite this

Akusok, A., Björk, K.-M., Lendasse, A., & Leal, L. E. (2026). Work-in-Progress: Evaluating Feasibility of Band Matrix Solvers for Scaling up Extreme Learning Machine Method. In Smart Technologies for an All-Electric Society: Proceedings of the 22nd International Conference on Smart Technologies & Education (STE2025) (Vol. 2, pp. 347-355). Springer. https://doi.org/10.1007/978-3-032-07319-8_31

@inbook{d3b2aa32b8d9419fb8b1e7f89df0ecf0,

title = "Work-in-Progress: Evaluating Feasibility of Band Matrix Solvers for Scaling up Extreme Learning Machine Method",

abstract = "This work presents the results of the potential of band linear system solvers for improving the scalability of the Extreme Learning Machine (ELM) method at large model sizes. The model is tested on the standard MNIST dataset with a range of solvers provided by the SciPy Python library. The results are analyzed taking into consideration the overall performance and the performance impact of band solvers across different matrix bandwidths, as well as the performance versus runtime analysis. The findings show potential in applying the proposed method to very large ELM models with narrow band matrices.",

author = "Anton Akusok and Kaj-Mikael Bj{\"o}rk and Amaury Lendasse and Leal, \{Leonardo Espinosa\}",

year = "2026",

doi = "10.1007/978-3-032-07319-8\_31",

language = "English",

isbn = "9783032073181",

volume = "2",

pages = "347--355",

booktitle = "Smart Technologies for an All-Electric Society",

publisher = "Springer",

address = "Germany",

}

Akusok, A, Björk, K-M, Lendasse, A & Leal, LE 2026, Work-in-Progress: Evaluating Feasibility of Band Matrix Solvers for Scaling up Extreme Learning Machine Method. in Smart Technologies for an All-Electric Society: Proceedings of the 22nd International Conference on Smart Technologies & Education (STE2025). vol. 2, Springer, pp. 347-355. https://doi.org/10.1007/978-3-032-07319-8_31

Work-in-Progress: Evaluating Feasibility of Band Matrix Solvers for Scaling up Extreme Learning Machine Method. / Akusok, Anton; Björk, Kaj-Mikael; Lendasse, Amaury et al.
Smart Technologies for an All-Electric Society: Proceedings of the 22nd International Conference on Smart Technologies & Education (STE2025). Vol. 2 Springer, 2026. p. 347-355.

Research output: Chapter in Book/Report/Conference proceeding › Chapter or book article › Scientific › peer-review

TY - CHAP

T1 - Work-in-Progress: Evaluating Feasibility of Band Matrix Solvers for Scaling up Extreme Learning Machine Method

AU - Akusok, Anton

AU - Björk, Kaj-Mikael

AU - Lendasse, Amaury

AU - Leal, Leonardo Espinosa

PY - 2026

Y1 - 2026

N2 - This work presents the results of the potential of band linear system solvers for improving the scalability of the Extreme Learning Machine (ELM) method at large model sizes. The model is tested on the standard MNIST dataset with a range of solvers provided by the SciPy Python library. The results are analyzed taking into consideration the overall performance and the performance impact of band solvers across different matrix bandwidths, as well as the performance versus runtime analysis. The findings show potential in applying the proposed method to very large ELM models with narrow band matrices.

AB - This work presents the results of the potential of band linear system solvers for improving the scalability of the Extreme Learning Machine (ELM) method at large model sizes. The model is tested on the standard MNIST dataset with a range of solvers provided by the SciPy Python library. The results are analyzed taking into consideration the overall performance and the performance impact of band solvers across different matrix bandwidths, as well as the performance versus runtime analysis. The findings show potential in applying the proposed method to very large ELM models with narrow band matrices.

U2 - 10.1007/978-3-032-07319-8_31

DO - 10.1007/978-3-032-07319-8_31

M3 - Chapter or book article

SN - 9783032073181

VL - 2

SP - 347

EP - 355

BT - Smart Technologies for an All-Electric Society

PB - Springer

ER -

Akusok A, Björk KM, Lendasse A, Leal LE. Work-in-Progress: Evaluating Feasibility of Band Matrix Solvers for Scaling up Extreme Learning Machine Method. In Smart Technologies for an All-Electric Society: Proceedings of the 22nd International Conference on Smart Technologies & Education (STE2025). Vol. 2. Springer. 2026. p. 347-355 doi: 10.1007/978-3-032-07319-8_31

Work-in-Progress: Evaluating Feasibility of Band Matrix Solvers for Scaling up Extreme Learning Machine Method

Abstract

Access to Document

Fingerprint

Cite this