Ändra sökning
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Energy-Aware Very Fast Decision Tree
Blekinge Tekniska Högskola, Fakulteten för datavetenskaper, Institutionen för datavetenskap.ORCID-id: 0000-0003-4973-9255
Blekinge Tekniska Högskola, Fakulteten för datavetenskaper, Institutionen för datavetenskap.ORCID-id: 0000-0002-0535-1761
Blekinge Tekniska Högskola, Fakulteten för datavetenskaper, Institutionen för datavetenskap.ORCID-id: 0000-0001-9947-1088
Blekinge Tekniska Högskola, Fakulteten för datavetenskaper, Institutionen för datavetenskap.ORCID-id: 0000-0002-3118-5058
Visa övriga samt affilieringar
2021 (Engelska)Ingår i: International Journal of Data Science and Analytics, ISSN 2364-415X, Vol. 11, nr 2, s. 105-126Artikel i tidskrift (Refereegranskat) Published
Abstract [en]

Recently machine learning researchers are designing algorithms that can run in embedded and mobile devices, which introduces additional constraints compared to traditional algorithm design approaches. One of these constraints is energy consumption, which directly translates to battery capacity for these devices. Streaming algorithms, such as the Very Fast Decision Tree (VFDT), are designed to run in such devices due to their high velocity and low memory requirements. However, they have not been designed with an energy efficiency focus. This paper addresses this challenge by presenting the nmin adaptation method, which reduces the energy consumption of the VFDT algorithm with only minor effects on accuracy. nmin adaptation allows the algorithm to grow faster in those branches where there is more confidence to create a split, and delays the split on the less confident branches. This removes unnecessary computations related to checking for splits but maintains similar levels of accuracy. We have conducted extensive experiments on 29 public datasets, showing that the VFDT with nmin adaptation consumes up to 31% less energy than the original VFDT, and up to 96% less energy than the CVFDT (VFDT adapted for concept drift scenarios), trading off up to 1.7 percent of accuracy.

Ort, förlag, år, upplaga, sidor
Springer Science+Business Media B.V., 2021. Vol. 11, nr 2, s. 105-126
Nationell ämneskategori
Datavetenskap (datalogi)
Identifikatorer
URN: urn:nbn:se:bth-19150DOI: 10.1007/s41060-021-00246-4ISI: 000631559600001Scopus ID: 2-s2.0-85102938796OAI: oai:DiVA.org:bth-19150DiVA, id: diva2:1388159
Ingår i projekt
Bigdata@BTH- Scalable resource-efficient systems for big data analytics, KK-stiftelsen
Forskningsfinansiär
KK-stiftelsen, 20140032
Anmärkning

open access

Tillgänglig från: 2020-01-23 Skapad: 2020-01-23 Senast uppdaterad: 2025-09-30Bibliografiskt granskad
Ingår i avhandling
1. Energy Efficiency in Machine Learning: Approaches to Sustainable Data Stream Mining
Öppna denna publikation i ny flik eller fönster >>Energy Efficiency in Machine Learning: Approaches to Sustainable Data Stream Mining
2020 (Engelska)Doktorsavhandling, sammanläggning (Övrigt vetenskapligt)
Abstract [en]

Energy efficiency in machine learning explores how to build machine learning algorithms and models with low computational and power requirements. Although energy consumption is starting to gain interest in the field of machine learning, still the majority of solutions focus on obtaining the highest predictive accuracy, without a clear focus on sustainability.

This thesis explores green machine learning, which builds on green computing and computer architecture to design sustainable and energy efficient machine learning algorithms. In particular, we investigate how to design machine learning algorithms that automatically learn from streaming data in an energy efficient manner.

We first illustrate how energy can be measured in the context of machine learning, in the form of a literature review and a procedure to create theoretical energy models. We use this knowledge to analyze the energy footprint of Hoeffding trees, presenting an energy model that maps the number of computations and memory accesses to the main functionalities of the algorithm. We also analyze the hardware events correlated to the execution of the algorithm, their functions and their hyper parameters.

The final contribution of the thesis is showcased by two novel extensions of Hoeffding tree algorithms, the Hoeffding tree with nmin adaptation and the Green Accelerated Hoeffding Tree. These solutions are able to reduce their energy consumption by twenty and thirty percent, with minimal effect on accuracy. This is achieved by setting an individual splitting criteria for each branch of the decision tree, spending more energy on the fast growing branches and saving energy on the rest.

This thesis shows the importance of evaluating energy consumption when designing machine learning algorithms, proving that we can design more energy efficient algorithms and still achieve competitive accuracy results.

Ort, förlag, år, upplaga, sidor
Karlskrona: Blekinge Tekniska Högskola, 2020. s. 267
Serie
Blekinge Institute of Technology Doctoral Dissertation Series, ISSN 1653-2090 ; 2
Nyckelord
machine learning, energy efficiency, data stream mining, green machine learning, edge computing
Nationell ämneskategori
Datavetenskap (datalogi)
Forskningsämne
Datavetenskap
Identifikatorer
urn:nbn:se:bth-18986 (URN)978-91-7295-396-3 (ISBN)
Disputation
2020-01-31, J1650, Blekinge Institute of Technology, Karlskrona, 13:15 (Engelska)
Opponent
Handledare
Forskningsfinansiär
KK-stiftelsen, 20140032
Tillgänglig från: 2019-12-03 Skapad: 2019-12-03 Senast uppdaterad: 2025-09-30Bibliografiskt granskad

Open Access i DiVA

fulltext(1437 kB)366 nedladdningar
Filinformation
Filnamn FULLTEXT01.pdfFilstorlek 1437 kBChecksumma SHA-512
9387e7ad91c0fd224a406901966942fdc42dcfcf47964022522453799608bef5516257f50f97389d635e946ec7df67bb73064a15244375e4f73e23e83bc8cfcc
Typ fulltextMimetyp application/pdf

Övriga länkar

Förlagets fulltextScopus

Person

García Martín, EvaLavesson, NiklasGrahn, HåkanCasalicchio, EmilianoBoeva, Veselka

Sök vidare i DiVA

Av författaren/redaktören
García Martín, EvaLavesson, NiklasGrahn, HåkanCasalicchio, EmilianoBoeva, Veselka
Av organisationen
Institutionen för datavetenskap
Datavetenskap (datalogi)

Sök vidare utanför DiVA

GoogleGoogle Scholar
Totalt: 367 nedladdningar
Antalet nedladdningar är summan av nedladdningar för alla fulltexter. Det kan inkludera t.ex tidigare versioner som nu inte längre är tillgängliga.

doi
urn-nbn

Altmetricpoäng

doi
urn-nbn
Totalt: 1163 träffar
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf