Javascript must be enabled for the correct page display

Many Layered Support Vector Machines

Asselt, J. van (2013) Many Layered Support Vector Machines. Bachelor's Thesis, Artificial Intelligence.

[img]
Preview
Text
AI_BA_2013_JEROENVANASSELT.pdf - Published Version

Download (322kB) | Preview
[img] Text
AkkoordWiering4.pdf - Other
Restricted to Repository staff only

Download (40kB)

Abstract

This thesis describes the expansion of the machine learning algorithm known as the two-layered Sup- port Vector Machine (SVM) (Wiering, Schutten, Millea, Meijster, and Schomaker, 2013) in order to improve the performance of this system. First, a de- scription of the standard SVM is given, followed by an overview of the two-layered SVM. Then another layer of SVMs is added to the two-layered SVM. The mathematical implications of this expansion will be described in detail. The metaparameters of the three-layered SVM are then found by using a combination of particle swarm optimisation and the UCB bandit algorithm. Experimental results show that the three-layered SVM outperforms a single- layered SVM. However, these results do not show an increase in performance compared to the two- layered SVM.

Item Type: Thesis (Bachelor's Thesis)
Degree programme: Artificial Intelligence
Thesis type: Bachelor's Thesis
Language: English
Date Deposited: 15 Feb 2018 07:54
Last Modified: 15 Feb 2018 07:54
URI: http://fse.studenttheses.ub.rug.nl/id/eprint/11248

Actions (login required)

View Item View Item