Standard

Star algorithm for neural network ensembling. / Zinchenko, Sergey; Lishudi, Dmitrii.

In: Neural networks : the official journal of the International Neural Network Society, Vol. 170, 02.2024, p. 364-375.

Research output: Contribution to journalArticlepeer-review

Harvard

Zinchenko, S & Lishudi, D 2024, 'Star algorithm for neural network ensembling', Neural networks : the official journal of the International Neural Network Society, vol. 170, pp. 364-375. https://doi.org/10.1016/j.neunet.2023.11.020

APA

Zinchenko, S., & Lishudi, D. (2024). Star algorithm for neural network ensembling. Neural networks : the official journal of the International Neural Network Society, 170, 364-375. https://doi.org/10.1016/j.neunet.2023.11.020

Vancouver

Zinchenko S, Lishudi D. Star algorithm for neural network ensembling. Neural networks : the official journal of the International Neural Network Society. 2024 Feb;170:364-375. doi: 10.1016/j.neunet.2023.11.020

Author

Zinchenko, Sergey ; Lishudi, Dmitrii. / Star algorithm for neural network ensembling. In: Neural networks : the official journal of the International Neural Network Society. 2024 ; Vol. 170. pp. 364-375.

BibTeX

@article{ceabd9bb5f834b67b1173c7474d1c283,
title = "Star algorithm for neural network ensembling",
abstract = "Neural network ensembling is a common and robust way to increase model efficiency. In this paper, we propose a new neural network ensemble algorithm based on Audibert's empirical star algorithm. We provide optimal theoretical minimax bound on the excess squared risk. Additionally, we empirically study this algorithm on regression and classification tasks and compare it to most popular ensembling methods.",
author = "Sergey Zinchenko and Dmitrii Lishudi",
note = "The publication was supported by the grant for research centers in the field of AI provided by the Analytical Center for the Government of the Russian Federation (ACRF) in accordance with the agreement on the provision of subsidies (identifier of the agreement 000000D730321P5Q0002) and the agreement with HSE University No. 70-2021-00139.We are grateful to Nikita Puchkin for essential comments and productive discussions, and also to Alexander Trushin for help with the design of the work. Copyright {\textcopyright} 2023 Elsevier Ltd. All rights reserved.",
year = "2024",
month = feb,
doi = "10.1016/j.neunet.2023.11.020",
language = "English",
volume = "170",
pages = "364--375",
journal = "Neural networks : the official journal of the International Neural Network Society",
issn = "0893-6080",
publisher = "Elsevier Ltd",

}

RIS

TY - JOUR

T1 - Star algorithm for neural network ensembling

AU - Zinchenko, Sergey

AU - Lishudi, Dmitrii

N1 - The publication was supported by the grant for research centers in the field of AI provided by the Analytical Center for the Government of the Russian Federation (ACRF) in accordance with the agreement on the provision of subsidies (identifier of the agreement 000000D730321P5Q0002) and the agreement with HSE University No. 70-2021-00139.We are grateful to Nikita Puchkin for essential comments and productive discussions, and also to Alexander Trushin for help with the design of the work. Copyright © 2023 Elsevier Ltd. All rights reserved.

PY - 2024/2

Y1 - 2024/2

N2 - Neural network ensembling is a common and robust way to increase model efficiency. In this paper, we propose a new neural network ensemble algorithm based on Audibert's empirical star algorithm. We provide optimal theoretical minimax bound on the excess squared risk. Additionally, we empirically study this algorithm on regression and classification tasks and compare it to most popular ensembling methods.

AB - Neural network ensembling is a common and robust way to increase model efficiency. In this paper, we propose a new neural network ensemble algorithm based on Audibert's empirical star algorithm. We provide optimal theoretical minimax bound on the excess squared risk. Additionally, we empirically study this algorithm on regression and classification tasks and compare it to most popular ensembling methods.

UR - https://www.scopus.com/record/display.uri?eid=2-s2.0-85178336619&origin=inward&txGid=cbeb1b6e37bc9a5fd3945eb6c2bac855

UR - https://www.mendeley.com/catalogue/6846bfaa-06cb-3f87-86df-04e6f1e6419f/

U2 - 10.1016/j.neunet.2023.11.020

DO - 10.1016/j.neunet.2023.11.020

M3 - Article

C2 - 38029718

VL - 170

SP - 364

EP - 375

JO - Neural networks : the official journal of the International Neural Network Society

JF - Neural networks : the official journal of the International Neural Network Society

SN - 0893-6080

ER -

ID: 59277803