Исмаилов Вугар Эльман
Основные научные достижения
1) Получены необходимые и достаточные условия для представления функций многих переменных линейными комбинациями ридж функций;
2) Доказана теорема чебышевского типа для экстремальности суммы ридж функций к заданной непрерывной функции;
3) Получены явные формулы для вычисления точного значения погрешности приближения и конструктивного построения наилучше приближающей функции в задачах равномерного и квадратичного приближения функций многих переменных суммами ридж функций;
4) Доказано, что если непрерывные функции, определенные на некотором компактном хаусдорфовом пространстве, представляются линейными суперпозициями, то всякая разрывная функция, определенная на этом пространстве также имеет такое представление.
5) решена задача теории аппроксимации функций многих переменных, связанная с теоремой Голомба.
Ряд результатов включены в книгу “Allan Pinkus, Ridge Functions, Cambridge University Press, 2015, 218 pp.” По некоторым результатам был сделан приглашенный доклад в Оксфордском университете (см. https://www.maths.ox.ac.uk/node/24710)
Список основных научных работ за последние пять лет
- (with N. Guliyev) On the approximation by single hidden layer feedforward neural networks with fixed weights, Neural Networks98(2018), 296-304, https://doi.org/10.1016/j.neunet.2017.12.007
- A note on the criterion for a best approximation by superpositions of functions, Studia Mathematica240 (2018), no. 2, 193-199, https://doi.org/10.4064/sm170314-9-4
- (with A. Asgarova) On the representation by sums of algebras of continuous functions, Comptes Rendus Mathematique355 (2017), no. 9, 949-955, https://doi.org/10.1016/j.crma.2017.09.015
-
A note on the equioscillation theorem for best ridge function approximation, Expositiones Mathematicae 35 (2017), no. 3, 343-349, https://doi.org/10.1016/j.exmath.2017.05.003
- (with A. Asgarova) Diliberto–Straus algorithm for the uniform approximation by a sum of two algebras, Proceedings – Mathematical Sciences127 (2017), no. 2, 361-374, http://dx.doi.org/10.1007/s12044-017-0337-4
- (with E. Savas) Measure theoretic results for approximation by neural networks with limited weights, Numerical Functional Analysis and Optimization38(2017), no. 7, 819-830, http://dx.doi.org/10.1080/01630563.2016.1254654
- Approximation by sums of ridge functions with fixed directions, (Russian) Algebra i Analiz28(2016), no. 6, 20–69, http://mi.mathnet.ru/eng/aa1513 English transl. St. Petersburg Mathematical Journal 28 (2017), 741-772, https://doi.org/10.1090/spmj/1471
- On the uniqueness of representation by linear superpositions, Ukrainskii Matematicheskii Zhurnal68(2016), no. 12, 1620-1628. English transl. Ukrainian Mathematical Journal 68 (2017), no. 12, 1874-1883, https://doi.org/10.1007/s11253-017-1335-5
- (with N. Guliyev) A single hidden layer feedforward network with only one neuron in the hidden layer can approximate any univariate function, Neural Computation 28(2016), no. 7, 1289–1304, http://dx.doi.org/10.1162/NECO_a_00849
- (with R. Aliev) On a smoothness problem in ridge function representation, Advances in Applied Mathematics73(2016), 154–169, http://dx.doi.org/10.1016/j.aam.2015.11.002
- Approximation by ridge functions and neural networks with a bounded number of neurons, Applicable Analysis94(2015), no. 11, 2245-2260, http://dx.doi.org/10.1080/00036811.2014.979809
- On the approximation by neural networks with bounded number of neurons in hidden layers, Journal of Mathematical Analysis and Applications417 (2014), no. 2, 963–969, http://dx.doi.org/10.1016/j.jmaa.2014.03.092
- (with A. Pinkus) Interpolation on lines by ridge functions, Journal of Approximation Theory175(2013), 91-113, http://dx.doi.org/10.1016/j.jat.2013.07.010
14. Approximation by neural networks with weights varying on a finite set of directions, Journal of Mathematical Analysis and Applications 389 (2012), Issue 1, 72-83, http://dx.doi.org/10.1016/j.jmaa.2011.11.037
- A note on the representation of continuous functions by linear superpositions, Expositiones Mathematicae30(2012), Issue 1, 96-101, http://dx.doi.org/10.1016/j.exmath.2011.07.005
- On the theorem of M Golomb, Proceedings – Mathematical Sciences119(2009), no. 1, 45-52, http://dx.doi.org/10.1007/s12044-009-0005-4
- On the representation by linear superpositions, Journal of Approximation Theory151(2008), Issue 2 , 113-125, http://dx.doi.org/10.1016/j.jat.2007.09.003
- On the approximation by compositions of fixed multivariate functions with univariate functions, Studia Mathematica183(2007), 117-126, http://dx.doi.org/10.4064/sm183-2-2
- On the best L₂ approximation by ridge functions, Applied Mathematics E-Notes, 7(2007), 71-76, http://www.math.nthu.edu.tw/~amen/
- Representation of multivariate functions by sums of ridge functions, Journal of Mathematical Analysis and Applications331(2007), Issue 1, 184-190, http://dx.doi.org/10.1016/j.jmaa.2006.08.076
- Characterization of an extremal sum of ridge functions, Journal of Computational and Applied Mathematics205(2007), Issue 1, 105-115, http://dx.doi.org/10.1016/j.cam.2006.04.043
- Methods for computing the least deviation from the sums of functions of one variable, (Russian) Sibirskii Matematicheskii Zhurnal47(2006), no. 5, 1076 -1082; translation in Siberian Mathematical Journal 47 (2006), no. 5, 883–888, http://dx.doi.org/10.1007/s11202-006-0097-3