PUBLICATION DATE

2014-02-07

AUTHORS

Jian Wang, Qin-wei Fan, Ye-tian Fan, Wei Wu, Wen-yu Yang

TITLE

A pruning algorithm with L 1/2 regularizer for extreme learning machine

ISSUE

2

VOLUME

15

ISSN (print)

1869-1951

ISSN (electronic)

1869-196X

ABSTRACT

Compared with traditional learning methods such as the back propagation (BP) method, extreme learning machine provides much faster learning speed and needs less human intervention, and thus has been widely used. In this paper we combine the L 1/2 regularization method with extreme learning machine to prune extreme learning machine. A variable learning coefficient is employed to prevent too large a learning increment. A numerical experiment demonstrates that a network pruned L 1/2 regularization has fewer hidden nodes but provides better performance than both the original network and the network pruned by L 2 regularization.

How to use: Click on a object to move its position. Double click to open its homepage. Right click to preview its contents.

Download the RDF metadata as:   json-ld nt turtle xml License info


38 TRIPLES      30 PREDICATES      37 URIs      20 LITERALS

Subject Predicate Object
1 articles:f37ae18c86867b339dc937df15c47387 sg:abstract Abstract Compared with traditional learning methods such as the back propagation (BP) method, extreme learning machine provides much faster learning speed and needs less human intervention, and thus has been widely used. In this paper we combine the L 1/2 regularization method with extreme learning machine to prune extreme learning machine. A variable learning coefficient is employed to prevent too large a learning increment. A numerical experiment demonstrates that a network pruned L 1/2 regularization has fewer hidden nodes but provides better performance than both the original network and the network pruned by L 2 regularization.
2 sg:articleType OriginalPaper
3 sg:coverYear 2014
4 sg:coverYearMonth 2014-02
5 sg:ddsId jzus.C1300197
6 sg:ddsIdJournalBrand 11714
7 sg:doi 10.1631/jzus.C1300197
8 sg:doiLink http://dx.doi.org/10.1631/jzus.C1300197
9 sg:hasContributingOrganization grid-institutes:grid.30055.33
10 grid-institutes:grid.35155.37
11 sg:hasContribution contributions:5612fc709a293aab46867a5d9986df63
12 contributions:adb120aabe9e8076fe7ce4c57189a7b1
13 contributions:cbf5e88d10725c6b7385c1c05c48458a
14 contributions:cd2f93588eca4fed7df3c927b431674f
15 contributions:e63579b39c1ceb39b02b075048626072
16 sg:hasFieldOfResearchCode anzsrc-for:08
17 anzsrc-for:0801
18 sg:hasJournal journals:e38e592a6525684cef9042f8fc0dbe7c
19 journals:fb867a50c83e333c24042ab9025c877b
20 sg:hasJournalBrand journal-brands:8fb2d2c7c9d0e7652ac272d1d08e3433
21 sg:indexingDatabase Scopus
22 Web of Science
23 sg:issnElectronic 1869-196X
24 sg:issnPrint 1869-1951
25 sg:issue 2
26 sg:language English
27 sg:license http://scigraph.springernature.com/explorer/license/
28 sg:pageEnd 125
29 sg:pageStart 119
30 sg:publicationDate 2014-02-07
31 sg:publicationYear 2014
32 sg:publicationYearMonth 2014-02
33 sg:scigraphId f37ae18c86867b339dc937df15c47387
34 sg:title A pruning algorithm with L 1/2 regularizer for extreme learning machine
35 sg:volume 15
36 sg:webpage https://link.springer.com/10.1631/jzus.C1300197
37 rdf:type sg:Article
38 rdfs:label Article: A pruning algorithm with L 1/2 regularizer for extreme learning machine
HOW TO GET THIS DATA PROGRAMMATICALLY:

JSON-LD is a popular JSON format for linked data.

curl -H 'Accept: application/ld+json' 'http://scigraph.springernature.com/things/articles/f37ae18c86867b339dc937df15c47387'

N-Triples is a line-based linked data format ideal for batch operations .

curl -H 'Accept: application/n-triples' 'http://scigraph.springernature.com/things/articles/f37ae18c86867b339dc937df15c47387'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'http://scigraph.springernature.com/things/articles/f37ae18c86867b339dc937df15c47387'

RDF/XML is a standard XML format for linked data.

curl -H 'Accept: application/rdf+xml' 'http://scigraph.springernature.com/things/articles/f37ae18c86867b339dc937df15c47387'






Preview window. Press ESC to close (or click here)


...