Ar gradientu saistītās minimizācijas metodes
Loading...
Date
Authors
Advisor
Journal Title
Journal ISSN
Volume Title
Publisher
Latvijas Universitāte
Language
lav
Abstract
Bakalaura darbs veltīts vairākargumentu funkciju optimizācijas metodēm, kuras sauc par gradienta metodēm. Darbā apskatītas četras metodes: gradienta metode, gradienta projekcijas metode, nosacītā gradienta metode un stohastiskā gradienta metode. Visām metodēm pievienoti piemēri, kas ilustrē metodes darbību. Metodiska rakstura darbs.
The Bachelors thesis is dedicated to multi variable function optimization methods, which are also known as gradient descent method. The work contains a description of 4 methods: gradient descent method, gradient projection method, conditional gradient method and stochastic descent gradient. Illustrative examples are given to describe each method in use. This is a methodical work.
The Bachelors thesis is dedicated to multi variable function optimization methods, which are also known as gradient descent method. The work contains a description of 4 methods: gradient descent method, gradient projection method, conditional gradient method and stochastic descent gradient. Illustrative examples are given to describe each method in use. This is a methodical work.