Extreme Learning Tree

Anton Akusok; Emil Eirola; Kaj-Mikael Björk; Amaury Lendasse

doi:10.1007/978-3-030-01520-6_16

Extreme Learning Tree

Anton Akusok, Emil Eirola, Kaj-Mikael Björk, Amaury Lendasse

Finansiell ekonomi, Helsingfors

Forskningsoutput: Kapitel i bok/rapport/konferenshandling › Konferensbidrag › Vetenskaplig › Peer review

Sammanfattning

The paper proposes a new variant of a decision tree, called an Extreme Learning Tree. It consists of an extremely random tree with non-linear data transformation, and a linear observer that provides predictions based on the leaf index where the data samples fall. The proposed method outperforms linear models on a benchmark dataset, and may be a building block for a future variant of Random Forest.

Originalspråk	Engelska
Titel på värdpublikation	International Conference on Extreme Learning Machine: Proceedings of ELM-2017
Antal sidor	5
Utgivningsort	Cham
Förlag	Springer
Utgivningsdatum	17.10.2018
Sidor	181-185
ISBN (tryckt)	978-3-030-01519-0
ISBN (elektroniskt)	978-3-030-01520-6
DOI	https://doi.org/10.1007/978-3-030-01520-6_16
Status	Publicerad - 17.10.2018
MoE-publikationstyp	A4 Artikel i en konferenspublikation
Evenemang	2017 the 8th International Conference on Extreme Learning Machines (ELM) - Yantai, Kina Varaktighet: 04.10.2017 → 07.10.2017 http://www.ntu.edu.sg/home/egbhuang/elm2017/index.html

Publikationsserier

Namn	Proceedings in Adaptation, Learning and Optimization (PALO)
Volym	10

Nyckelord

512 Företagsekonomi

Åtkomst av dokument

10.1007/978-3-030-01520-6_16

OpenUrl-tillgänglighet

Fulltext

Citera det här

@inproceedings{04d9fbcd9d724f86b7751b6386049ea5,

title = "Extreme Learning Tree",

abstract = "The paper proposes a new variant of a decision tree, called an Extreme Learning Tree. It consists of an extremely random tree with non-linear data transformation, and a linear observer that provides predictions based on the leaf index where the data samples fall. The proposed method outperforms linear models on a benchmark dataset, and may be a building block for a future variant of Random Forest.",

keywords = "512 Business and Management, ELM, Decision tree, Randomized methods",

author = "Anton Akusok and Emil Eirola and Kaj-Mikael Bj{\"o}rk and Amaury Lendasse",

year = "2018",

month = oct,

day = "17",

doi = "10.1007/978-3-030-01520-6_16",

language = "English",

isbn = "978-3-030-01519-0",

series = "Proceedings in Adaptation, Learning and Optimization (PALO)",

publisher = "Springer",

pages = "181--185 ",

booktitle = "International Conference on Extreme Learning Machine: Proceedings of ELM-2017",

address = "International",

note = "2017 the 8th International Conference on Extreme Learning Machines (ELM) , ELM-2017 ; Conference date: 04-10-2017 Through 07-10-2017",

url = "http://www.ntu.edu.sg/home/egbhuang/elm2017/index.html",

}

Akusok, A, Eirola, E, Björk, K-M & Lendasse, A 2018, Extreme Learning Tree. i International Conference on Extreme Learning Machine: Proceedings of ELM-2017. Proceedings in Adaptation, Learning and Optimization (PALO), vol. 10, Springer, Cham, s. 181-185 , 2017 the 8th International Conference on Extreme Learning Machines (ELM) , Yantai, Kina, 04.10.2017. https://doi.org/10.1007/978-3-030-01520-6_16

Extreme Learning Tree. / Akusok, Anton; Eirola, Emil; Björk, Kaj-Mikael et al.

International Conference on Extreme Learning Machine: Proceedings of ELM-2017. Cham : Springer, 2018. s. 181-185 (Proceedings in Adaptation, Learning and Optimization (PALO); Vol. 10).

Forskningsoutput: Kapitel i bok/rapport/konferenshandling › Konferensbidrag › Vetenskaplig › Peer review

TY - GEN

T1 - Extreme Learning Tree

AU - Akusok, Anton

AU - Eirola, Emil

AU - Björk, Kaj-Mikael

AU - Lendasse, Amaury

PY - 2018/10/17

Y1 - 2018/10/17

N2 - The paper proposes a new variant of a decision tree, called an Extreme Learning Tree. It consists of an extremely random tree with non-linear data transformation, and a linear observer that provides predictions based on the leaf index where the data samples fall. The proposed method outperforms linear models on a benchmark dataset, and may be a building block for a future variant of Random Forest.

AB - The paper proposes a new variant of a decision tree, called an Extreme Learning Tree. It consists of an extremely random tree with non-linear data transformation, and a linear observer that provides predictions based on the leaf index where the data samples fall. The proposed method outperforms linear models on a benchmark dataset, and may be a building block for a future variant of Random Forest.

KW - 512 Business and Management

KW - ELM

KW - Decision tree

KW - Randomized methods

U2 - 10.1007/978-3-030-01520-6_16

DO - 10.1007/978-3-030-01520-6_16

M3 - Conference contribution

SN - 978-3-030-01519-0

T3 - Proceedings in Adaptation, Learning and Optimization (PALO)

SP - 181

EP - 185

BT - International Conference on Extreme Learning Machine: Proceedings of ELM-2017

PB - Springer

CY - Cham

T2 - 2017 the 8th International Conference on Extreme Learning Machines (ELM)

Y2 - 4 October 2017 through 7 October 2017

ER -

Extreme Learning Tree

Sammanfattning

Publikationsserier

Nyckelord

Åtkomst av dokument

OpenUrl-tillgänglighet

Fingeravtryck

Citera det här