Stochastic gradient algorithm based on the average aggregate functions
Vestnik KRAUNC. Fiziko-matematičeskie nauki, no. 5 (2016), pp. 112-125
Voir la notice de l'article provenant de la source Math-Net.Ru
The paper proposes a new scheme for the gradient solution to minimize losses averaged problem. It is an analog circuit used in the SAG algorithm in the case when the risk is calculated using the arithmetic mean. An illustrative example of the construction of robust classification based on the maximization of the surrogate median indentation.
Keywords:
Empirical risk, classification problem, averaging aggregation function, gradient based algorithm.
@article{VKAM_2016_5_a16,
author = {Z. M. Shibzukhov and M. A. Kazakov},
title = {Stochastic gradient algorithm based on the average aggregate functions},
journal = {Vestnik KRAUNC. Fiziko-matemati\v{c}eskie nauki},
pages = {112--125},
publisher = {mathdoc},
number = {5},
year = {2016},
language = {ru},
url = {http://geodesic.mathdoc.fr/item/VKAM_2016_5_a16/}
}
TY - JOUR AU - Z. M. Shibzukhov AU - M. A. Kazakov TI - Stochastic gradient algorithm based on the average aggregate functions JO - Vestnik KRAUNC. Fiziko-matematičeskie nauki PY - 2016 SP - 112 EP - 125 IS - 5 PB - mathdoc UR - http://geodesic.mathdoc.fr/item/VKAM_2016_5_a16/ LA - ru ID - VKAM_2016_5_a16 ER -
Z. M. Shibzukhov; M. A. Kazakov. Stochastic gradient algorithm based on the average aggregate functions. Vestnik KRAUNC. Fiziko-matematičeskie nauki, no. 5 (2016), pp. 112-125. http://geodesic.mathdoc.fr/item/VKAM_2016_5_a16/