kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Evaluating Brain-Inspired Machine Learning Models for Time Series Forecasting: A Comparative Study on Dynamical Memory in Reservoir Computing and Neural Networks
KTH, School of Electrical Engineering and Computer Science (EECS).
KTH, School of Electrical Engineering and Computer Science (EECS).
2023 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
Abstract [en]

Brain-inspired computing is a promising research field, with potential to encouragebreakthroughs within machine learning and enable us to solve complex problems in a moreefficient way. This study aims to compare the performance of brain-like machine learningalgorithms for time series forecasting. Three models were implemented: a vanilla recurrentneural network (VRNN), a more brain-like reservoir computing (RC) model, as well as a timelagged version of the latter (TLRC). Additionally, an autoregressive integrated movingaverage (ARIMA) was implemented to obtain benchmark results, since this is a well-established model with no connection to the brain. The performances were evaluated on aspectrum of univariate and multivariate time series, ranging from chaotic benchmark data toexperimental data, such as temperature recordings. The results indicate that the reservoircomputing models generally outperform the recurrent neural network, and that the consideredmodels have a disadvantage in practical scenarios. A common factor for these algorithms isthat they exhibit a sense of dynamical memory. Hence, we compare these models not only interms of their predictive accuracy, but also in terms of their capability for memory. To reachan analysis of dynamical memory, this study investigated the usefulness of the algorithms onvarying amounts of available information. Lastly, the effect of the network/reservoir size wastaken into account. The research suggests that the models cannot benefit meaningfully fromhaving more neurons.

Abstract [sv]

Beräkningsalgoritmer som är inspirerade av hjärnan har stor potential att bidra till genombrottinom maskininlärning och möjliggöra effektivare lösningar till komplexa problem. Dennastudie syftar till att jämföra prestandan hos beräkningsalgoritmer som liknar hjärnan,tillämpade för tidsserieprediktion. Tre algoritmer implementerades: en standardimplementerat återkommande neuralt nätverk (VRNN), en reservoir computing (RC) modelsom efterliknar hjärnan och en tidsfördröjd version av den senare (TLRC). Utöver dessaanvändes även en autoregressive integrated moving average (ARIMA) modell, för attgenerera referensresultat, då modellen är väletablerad och saknar anknytning till hjärnan.Prestandan evaluerades på ett spektrum av en variabla- och multivariabla tidsserier, frånkaotiska referensdata till experimentella data, såsom temperaturmätningar. Resultatenindikerar att reservoir computing modellerna generellt sett överträffar det återkommandeneurala nätverket, samt att praktiska scenarier är ogynnsamma för de betraktade modellerna.En gemensam nämnare för dessa algoritmer är att de besitter en typ av dynamiskt minne.Således jämfördes algoritmer inte enbart med hänsyn till deras prediktiva precision, men äveni termer av minneskapacitet. För att nå en analys av dynamiskt minne undersökte denna studiede respektive modellernas kapacitet för varierande mängder av tillgänglig information. Till sist beaktades inverkan av nätverkets/reservoarens storlek. Resultaten tyder på att modellernasprestanda inte meningsfullt kunde dra nytta av ett större antal neuroner.

Place, publisher, year, edition, pages
2023. , p. 691-704
Series
TRITA-EECS-EX ; 2023:197
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
URN: urn:nbn:se:kth:diva-341788OAI: oai:DiVA.org:kth-341788DiVA, id: diva2:1823500
Supervisors
Examiners
Projects
Kandidatexjobb i elektroteknik 2023, KTH, StockholmAvailable from: 2024-01-02 Created: 2024-01-02

Open Access in DiVA

fulltext(211487 kB)629 downloads
File information
File name FULLTEXT01.pdfFile size 211487 kBChecksum SHA-512
69786101c351a58f7bd524c3aeee40c661028b577366c4a725033372b88c624c87c2183b6acca2d3d43bbd2bb2f3942326c69263e70c99cf1db027ce9c4e9ae2
Type fulltextMimetype application/pdf

By organisation
School of Electrical Engineering and Computer Science (EECS)
Electrical Engineering, Electronic Engineering, Information Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 629 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

urn-nbn

Altmetric score

urn-nbn
Total: 491 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf