Bibliographic citations
This is an automatically generated citacion. Modify it if you see fit
Guimaray, H., (1989). Optimización no diferenciable [Tesis, Universidad Nacional de Ingeniería]. http://hdl.handle.net/20.500.14076/335
Guimaray, H., Optimización no diferenciable [Tesis]. : Universidad Nacional de Ingeniería; 1989. http://hdl.handle.net/20.500.14076/335
@mastersthesis{renati/703635,
title = "Optimización no diferenciable",
author = "Guimaray Huerta, Héctor Carlos",
publisher = "Universidad Nacional de Ingeniería",
year = "1989"
}
Title: Optimización no diferenciable
Authors(s): Guimaray Huerta, Héctor Carlos
Advisor(s): Blum R., Eugen
Keywords: Funciones no diferenciables; Matemática
Issue Date: 1989
Institution: Universidad Nacional de Ingeniería
Abstract: Muchos problemas de optimización de Investigación operativa, computación, ingeniería y ciencias no son regulares, esto es: incluyen funciones no diferenciables. Estos problemas no se dejan resol ver por los algoritmos clásicos de la optimización matemática, que fueron elaborados a partir de 1950.
La necesidad de crear nuevos métodos, para resolver estos problemas, condujo al desarrollo del campo de la optimización no diferenciable, que es muy reciente y su conocimiento se inicia entre 1975 y 1980.
Este trabajo se ha realizado en el contexto general de los espacios topológicos vectoriales dándose un sólido conocimiento de los aspectos principales del análisis convexo.
En el caso de funciones convexas el concepto de subdiferenciabilidad constituye una generalización de diferenciabilidad, para cuya ilustración se muestran los subdiferenciales de algunas funciones.
Consideramos las funciones conjugadas que proporcionan una útil herramienta para investigar propiedades de conjuntos convexos y funciones convexas, presentándose así la teoría de dualidad y sus aplicaciones.
Posteriormente, tratamos la continuidad superior de la correspondencia ∂Ef, la cual nos permite presentar el método de planos cortantes con subgradientes.
Finalmente, deseo expresar mi agradecimiento al Dr. Eugen Blum R. por el asesoramiento dado, igualmente al CONCYTEC por el apoyo económico brindado, contribuyendo así al desarrollo de la ciencia en nuestro país.
Many problems of operative investigation optimi¬zation computation, engineering and science are not regulars, that is: include no differentiable functions Those problems don't allowi ng themselves to sol ve for the classic algorithms of the mathematic optimi zation, that were elaborate since 1950. The necessity of creating new methods, for solving those problems, conduced to the development of the no differentiable optimization field, that is very recent and its knowledge begins between 1975 and 1980. This work has been done in the general context of the vector topological spaces with a sol id knowledge of the main aspects of the convex analysis. In the case of convex functions the concept of subdifferentiability constitutes a generalization of differentiabi1ity, for whose ilustration we have shown the subdifferentials of some functions. We consider the conjugate functions that provide a useful tool for investigating properties of convex sets and convex functions, which gives so the dua1ity theory and its applications. Later, we treat the upper continuity of the correspondence ∂Ef, which allows us to give the cutting hyperplane method with subgradients. Finally, I wish to express my gratitude to Dr. Eugen Blum for suggestions constructive and a careful reading of manuscript. Thanks are also due to CONCYTEC for the economic support, contributing so to the development of the science in our country.
Many problems of operative investigation optimi¬zation computation, engineering and science are not regulars, that is: include no differentiable functions Those problems don't allowi ng themselves to sol ve for the classic algorithms of the mathematic optimi zation, that were elaborate since 1950. The necessity of creating new methods, for solving those problems, conduced to the development of the no differentiable optimization field, that is very recent and its knowledge begins between 1975 and 1980. This work has been done in the general context of the vector topological spaces with a sol id knowledge of the main aspects of the convex analysis. In the case of convex functions the concept of subdifferentiability constitutes a generalization of differentiabi1ity, for whose ilustration we have shown the subdifferentials of some functions. We consider the conjugate functions that provide a useful tool for investigating properties of convex sets and convex functions, which gives so the dua1ity theory and its applications. Later, we treat the upper continuity of the correspondence ∂Ef, which allows us to give the cutting hyperplane method with subgradients. Finally, I wish to express my gratitude to Dr. Eugen Blum for suggestions constructive and a careful reading of manuscript. Thanks are also due to CONCYTEC for the economic support, contributing so to the development of the science in our country.
Link to repository: http://hdl.handle.net/20.500.14076/335
Discipline: Maestría en Ciencias con Mención en Matemática
Grade or title grantor: Universidad Nacional de Ingeniería. Facultad de Ciencias. Unidad de Posgrado
Grade or title: Maestro en Ciencias con Mención en Matemática
Register date: 4-Sep-2013
This item is licensed under a Creative Commons License