1984, vol. 8, núm. 4
http://hdl.handle.net/2099/3830
2024-03-28T22:17:06ZÍndex
http://hdl.handle.net/2099/5577
Índex
Qüestiió
2008-06-12T17:30:11ZQüestiióAn isolated word recognition system based on a low-complexity parametrization procedure
http://hdl.handle.net/2099/4399
An isolated word recognition system based on a low-complexity parametrization procedure
Rulot Segovia, Héctor; Vidal Ruiz, Enrique; Casacuberta Nolla, Francisco
An Isolated Word Recognition System is presented in this paper which uses a parametrization scheme based on the two-level clipped signal Autocorrelation Function. The system prototype runs on a 64 kby. partition of a general-purpose minicomputer with quite small specific hardware requirements and, for moderate sized dictionaries (≤ 40 words), gives 95-98% recognition rates with response times better than two times real time. The system uses claasical Dynamic Programming word-matching, and its main-aimed applications are the implementation of low-cost microprocessor-based speech devices and the incorporation of Isolated Word Recognizers among the software utilities of general-purpose (mini-micro) computers. The main system features are discussed and formal evaluation tests are presented.
2008-02-28T19:44:52ZRulot Segovia, HéctorVidal Ruiz, EnriqueCasacuberta Nolla, FranciscoAn Isolated Word Recognition System is presented in this paper which uses a parametrization scheme based on the two-level clipped signal Autocorrelation Function. The system prototype runs on a 64 kby. partition of a general-purpose minicomputer with quite small specific hardware requirements and, for moderate sized dictionaries (≤ 40 words), gives 95-98% recognition rates with response times better than two times real time. The system uses claasical Dynamic Programming word-matching, and its main-aimed applications are the implementation of low-cost microprocessor-based speech devices and the incorporation of Isolated Word Recognizers among the software utilities of general-purpose (mini-micro) computers. The main system features are discussed and formal evaluation tests are presented.Some aspects of parameter inference for nearly nonstationary and nearly noninvertible ARMA Models II
http://hdl.handle.net/2099/3918
Some aspects of parameter inference for nearly nonstationary and nearly noninvertible ARMA Models II
Ahtola, Juha; Tiao, George C.
This article will extend the discussion in Ahtola and Tiao (1984a) of the finite sample distribution of the score function in nearly nonstationary first order autoregressions to nearly noninvertible first order moving average models. This distribution theory can be used to appreciate the behavior of the score function in situations where the asymptotic normal theory is known to give poor approximations in finite samples.
The approximate distributions suggested here can be used to test for the value of the moving average parameter when it is close to unity. In particular, a test for noninvertibility can be obtained with an exact finite sample distribution of the test statistic under the null hypothesis.
2007-11-29T17:09:54ZAhtola, JuhaTiao, George C.This article will extend the discussion in Ahtola and Tiao (1984a) of the finite sample distribution of the score function in nearly nonstationary first order autoregressions to nearly noninvertible first order moving average models. This distribution theory can be used to appreciate the behavior of the score function in situations where the asymptotic normal theory is known to give poor approximations in finite samples.
The approximate distributions suggested here can be used to test for the value of the moving average parameter when it is close to unity. In particular, a test for noninvertibility can be obtained with an exact finite sample distribution of the test statistic under the null hypothesis.Revisión crítica de la utilización del método del gradiente conjugado y extensiones en la minimización de funciones nolineales con condiciones
http://hdl.handle.net/2099/3917
Revisión crítica de la utilización del método del gradiente conjugado y extensiones en la minimización de funciones nolineales con condiciones
Escudero, L. F.
En este trabajo se efectúa una revisión de la aplicación del método del gradiente conjugado (diseñado para resolver sistemas de ecuaciones lineales) a la optimización de funciones no lineales (no cuadráticas) en el conjunto factible definido por condiciones lineales. Se describen las desventajas de su utilización; se sugiere su sustitución por algoritmos que, precisando parecida capacidad de memoria, no adolecen de sus inconvenientes. Los resultados computacionales que se acompañan avalan esta sustitución; This paper undertakes a review of the application of the conjugate gradient method (designed to solve linear equation systems) to the optimization of non-linear functions (non-quadratic) in the feasible whole defined by linear conditions. The disadvantages of its use are outlined and its replacement by algorithms which, although requiring similar memory capacity, do not suffer from its drawbacks, is proposed. The computerized results support this replacement.
2007-11-29T17:07:51ZEscudero, L. F.En este trabajo se efectúa una revisión de la aplicación del método del gradiente conjugado (diseñado para resolver sistemas de ecuaciones lineales) a la optimización de funciones no lineales (no cuadráticas) en el conjunto factible definido por condiciones lineales. Se describen las desventajas de su utilización; se sugiere su sustitución por algoritmos que, precisando parecida capacidad de memoria, no adolecen de sus inconvenientes. Los resultados computacionales que se acompañan avalan esta sustitución
This paper undertakes a review of the application of the conjugate gradient method (designed to solve linear equation systems) to the optimization of non-linear functions (non-quadratic) in the feasible whole defined by linear conditions. The disadvantages of its use are outlined and its replacement by algorithms which, although requiring similar memory capacity, do not suffer from its drawbacks, is proposed. The computerized results support this replacement.Cotas inferiores para el problema de secuenciación con restricciones sobre los recursos
http://hdl.handle.net/2099/3916
Cotas inferiores para el problema de secuenciación con restricciones sobre los recursos
Álvarez-Valdés Olaguíbel, Ramón; Tamarit Goerlich, José Manuel
El trabajo explora dos vías de obtención de cotas inferiores para el problema de secuenciación de actividades con restricciones sobre los recursos, a partir de una formulación entera del problema. Una primera cota se obtiene de la relajación lineal y la aplicación sucesiva de planos de corte. El segundo método utiliza la relajación lagrangiana. El problema relajado se descompone en dos subproblemas para los que se proponen algoritmos de resolución. Se incluyen resultados computacionales que ilustran el comportamiento de las cotas obtenidas en ambos casos sobre una colección de problemas test.; The paper examines two means of obtaining lower bounds for the problem of sequencing activities with resource restrictions, from the starting point of an overall formulation of the problem. A first bound is obtained from the linear relaxation and the successive application of cutting planes. The second method uses Lagrangian relaxation. After applying relaxation analysis, the problem breaks down into two subproblems for which resolution algorithms are proposed. Computerized results are included, illustrating the behaviour of the bounds obtained in both cases on a set of test problems.
2007-11-29T16:55:24ZÁlvarez-Valdés Olaguíbel, RamónTamarit Goerlich, José ManuelEl trabajo explora dos vías de obtención de cotas inferiores para el problema de secuenciación de actividades con restricciones sobre los recursos, a partir de una formulación entera del problema. Una primera cota se obtiene de la relajación lineal y la aplicación sucesiva de planos de corte. El segundo método utiliza la relajación lagrangiana. El problema relajado se descompone en dos subproblemas para los que se proponen algoritmos de resolución. Se incluyen resultados computacionales que ilustran el comportamiento de las cotas obtenidas en ambos casos sobre una colección de problemas test.
The paper examines two means of obtaining lower bounds for the problem of sequencing activities with resource restrictions, from the starting point of an overall formulation of the problem. A first bound is obtained from the linear relaxation and the successive application of cutting planes. The second method uses Lagrangian relaxation. After applying relaxation analysis, the problem breaks down into two subproblems for which resolution algorithms are proposed. Computerized results are included, illustrating the behaviour of the bounds obtained in both cases on a set of test problems.Una aplicación de la teoría de Dubovickii y Miljutin a la programación semi-infinita convexa
http://hdl.handle.net/2099/3915
Una aplicación de la teoría de Dubovickii y Miljutin a la programación semi-infinita convexa
López Cerdá, Marco A.; Vercher, Enriqueta
En este trabajo aplicamos la teoría de Dubovickii y Miljutin para deducir una condición necesaria de optimalidad relativa al problema de Programación Semi-Infinita convexa no diferenciable, asumiendo la cualificación de Slater. Se introduce así un nuevo procedimiento para verificar la validez de esta cualificación.; In this paper we apply the Dubovickii and Miljutin theory in order to obtain a necessary condition for the relative optimality of the problem of semi-infinite convex non diferentiable programming by assuming the Slater's qualification. A new procedure is introduced for verifying the latter qualification.
2007-11-29T16:52:38ZLópez Cerdá, Marco A.Vercher, EnriquetaEn este trabajo aplicamos la teoría de Dubovickii y Miljutin para deducir una condición necesaria de optimalidad relativa al problema de Programación Semi-Infinita convexa no diferenciable, asumiendo la cualificación de Slater. Se introduce así un nuevo procedimiento para verificar la validez de esta cualificación.
In this paper we apply the Dubovickii and Miljutin theory in order to obtain a necessary condition for the relative optimality of the problem of semi-infinite convex non diferentiable programming by assuming the Slater's qualification. A new procedure is introduced for verifying the latter qualification.