Change search
ReferencesLink to record
Permanent link

Direct link
System Identification by Adaptive Boosting
KTH, School of Electrical Engineering (EES), Automatic Control.
2015 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

In the field of machine learning, the algorithm Adaptive Boosting has beensuccessfully applied to a wide range of regression and classification problems.Still, there is no known method to use the algorithm to estimate dynamical systems.In this thesis, the relationship between Adaptive Boosting and systemidentification is explored. A new identification method, inspired by AdaptiveBoosting, called TM-Boost is introduced. It fits a dynamical model byiteratively adding orthonormal basis functions. An interesting feature of themethod is that there is no need to specify a model order. It is also proven mathematicallyand verified in a series of identification experiments that TM-Boost,under reasonable conditions, converges to the true underlying system.

Abstract [sv]

Inom området maskininlärning har algoritmen Adaptive Boosting framgångs-rikt använts på många typer av klassificerings- och regressionsproblem. Hit-intills har algoritmen dock inte använts till att estimera dynamiska system. I detta examensarbete utforskas sambanden mellan Adaptive Boosting och sys-temidentifiering. En ny identifieringsmetod kallad TM-Boost, som är inspir-erad av Adaptive Boosting introduceras. Metoden baseras på ortonormala basfunktioner och bygger iterativt upp ett dynamiskt system. En tilltalande egenskap är att det inte längre är nödvändigt att specifiera modellordning. Det bevisas också matematiskt att det estimerade systemet, under vissa förut-sättningar, konvergerar mot det sanna underliggande systemet, vilket även verifieras i en serie identifieringsexperiment.

Place, publisher, year, edition, pages
2015. , 60 p.
EES Examensarbete / Master Thesis, TRITA EE 2015:70
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
URN: urn:nbn:se:kth:diva-179711OAI: diva2:885999
Available from: 2015-12-21 Created: 2015-12-21 Last updated: 2015-12-21Bibliographically approved

Open Access in DiVA

fulltext(1215 kB)61 downloads
File information
File name FULLTEXT01.pdfFile size 1215 kBChecksum SHA-512
Type fulltextMimetype application/pdf

By organisation
Automatic Control
Electrical Engineering, Electronic Engineering, Information Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 61 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 400 hits
ReferencesLink to record
Permanent link

Direct link