Abstract
A scheme for constructing a combination of estimation algorithms that is correct on a given training sample is proposed. The basic elements of the combination are algorithms obtained using the cross validation training technique.
Similar content being viewed by others
References
Yu. I. Zhuravlev, “Correct algorithms over sets of incorrect (heuristic) algorithms: Part I,” Kibernetika, No. 1, 5–17 (1977).
Yu. I. Zhuravlev, “Correct algorithms over sets of incorrect (heuristic) algorithms: Part II,” Kibernetika, No. 6, 21–27 (1977).
Yu. I. Zhuravlev, “Correct algorithms over sets of incorrect (heuristic) algorithms: Part III,” Kibernetika, No. 2, 35–43 (1978).
I. V. Isaev and Yu. I. Zhuravlev, “Designing classification algorithms that are correct for a given test sample,” Zh. Vychisl. Mat. Mat. Fiz. 19 (3), 726–738 (1979).
V. L. Matrosov, “On the completeness criteria for the model of estimation algorithms and its algebraic closure,” Dokl. Akad. Nauk SSSR 258 (4), 791–796 (1981).
A. G. D’yakonov, “Theory of equivalence systems for describing the algebraic closures of a generalized estimation model,” Zh. Vychisl. Mat. Mat. Fiz. 50, Comput. Math. Math. Phys. 50, 369–381 (2010).
A. G. D’yakonov, “Theory of equivalence systems for describing the algebraic closures of a generalized estimation model. II,” Comput. Math. Math. Phys. 51 (3), 490–504 (2011).
A. G. D’yakonov, “Investigation of algebraic closures of recognition algorithms: Labeling operators,” Tavrich. Vestn. Inf. Mat. 1, 199–203 (2008).
Author information
Authors and Affiliations
Corresponding author
Additional information
Original Russian Text © O.A. Ignat’ev, 2015, published in Zhurnal Vychislitel’noi Matematiki i Matematicheskoi Fiziki, 2015, Vol. 55, No. 12, pp. 2123–2129.
Rights and permissions
About this article
Cite this article
Ignat’ev, O.A. Construction of a correct combination of estimation algorithms adjusted using the cross validation technique. Comput. Math. and Math. Phys. 55, 2094–2099 (2015). https://doi.org/10.1134/S0965542515120064
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S0965542515120064