Title :
Global extremum search on the basis of density and its mode estimation
Author :
Vladimirova, Ludmila V.
Author_Institution :
St.-Peterburg State Univ., St. Petersburg, Russia
fDate :
June 30 2014-July 4 2014
Abstract :
Nonparametric density estimates may be used as a basis for construction of global random search algorithms. The distribution density reflects the main peculiarities of objective function behavior. Consequently, the density mode may be used as approximate argument value, providing objective function maximum.
Keywords :
density measurement; random processes; distribution density; global extremum search; global random search algorithms; mode estimation; Covariance matrices; Educational institutions; Ellipsoids; Estimation; Linear programming; Monte Carlo methods; Vectors;
Conference_Titel :
Beam Dynamics and Optimization (BDO), 2014 20th International Workshop on
Conference_Location :
St. Petersburg
Print_ISBN :
978-1-4799-5319-6
DOI :
10.1109/BDO.2014.6890096
