Active Learning Based on Single-Hidden Layer Feed-Forward Neural Network

Ran WANG; Sam KWONG; Qingshan JIANG; Ka-Chun WONG

doi:10.1109/SMC.2015.377

Active Learning Based on Single-Hidden Layer Feed-Forward Neural Network

Ran WANG, Sam KWONG, Qingshan JIANG, Ka-Chun WONG

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research › peer-review

3 Citations (Scopus)

Abstract

In this paper, we propose two stream-based active learning algorithms for single-hidden layer feed-forward neural networks (SLFNs) trained by extreme learning machine (ELM). Uncertainty and inconsistency are adopted as two sample selection criteria. Uncertainty reflects the nondeterminacy of a sample among different decision classes, which is calculated by information entropy or Gini-index. Inconsistency reflects the disagreement of the sample between its conditional features and decision labels, which is calculated by the lower approximations in fuzzy rough sets. Experimental results demonstrate that inconsistency-based strategy is more effective than uncertainty based strategy for SLFNs under stream-based environment.

Original language	English
Title of host publication	Proceedings : 2015 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2015
Publisher	IEEE
Pages	2158-2163
Number of pages	6
ISBN (Electronic)	9781479986972
ISBN (Print)	9781479986965
DOIs	https://doi.org/10.1109/SMC.2015.377
Publication status	Published - 12 Jan 2016
Externally published	Yes
Event	IEEE International Conference on Systems, Man, and Cybernetics, SMC 2015 - Kowloon Tong, Hong Kong, China Duration: 9 Oct 2015 → 12 Oct 2015

Conference

Conference	IEEE International Conference on Systems, Man, and Cybernetics, SMC 2015
Abbreviated title	SMC 2015
Country/Territory	Hong Kong, China
City	Kowloon Tong
Period	9/10/15 → 12/10/15

Funding

National Natural Science Foundation of China under Grant 61402460

Keywords

Active Learning
Extreme Learning Machine
Inconsistency
SLFNs
Uncertainty

Access to Document

10.1109/SMC.2015.377

Cite this

@inproceedings{22178dc733bc458b9e1bb412606a77e5,

title = "Active Learning Based on Single-Hidden Layer Feed-Forward Neural Network",

abstract = "In this paper, we propose two stream-based active learning algorithms for single-hidden layer feed-forward neural networks (SLFNs) trained by extreme learning machine (ELM). Uncertainty and inconsistency are adopted as two sample selection criteria. Uncertainty reflects the nondeterminacy of a sample among different decision classes, which is calculated by information entropy or Gini-index. Inconsistency reflects the disagreement of the sample between its conditional features and decision labels, which is calculated by the lower approximations in fuzzy rough sets. Experimental results demonstrate that inconsistency-based strategy is more effective than uncertainty based strategy for SLFNs under stream-based environment.",

keywords = "Active Learning, Extreme Learning Machine, Inconsistency, SLFNs, Uncertainty",

author = "Ran WANG and Sam KWONG and Qingshan JIANG and Ka-Chun WONG",

year = "2016",

month = jan,

day = "12",

doi = "10.1109/SMC.2015.377",

language = "English",

isbn = "9781479986965",

pages = "2158--2163",

booktitle = "Proceedings : 2015 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2015",

publisher = "IEEE",

note = "IEEE International Conference on Systems, Man, and Cybernetics, SMC 2015, SMC 2015 ; Conference date: 09-10-2015 Through 12-10-2015",

}

WANG, R, KWONG, S, JIANG, Q & WONG, K-C 2016, Active Learning Based on Single-Hidden Layer Feed-Forward Neural Network. in Proceedings : 2015 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2015. IEEE, pp. 2158-2163, IEEE International Conference on Systems, Man, and Cybernetics, SMC 2015, Kowloon Tong, Hong Kong, China, 9/10/15. https://doi.org/10.1109/SMC.2015.377

TY - GEN

T1 - Active Learning Based on Single-Hidden Layer Feed-Forward Neural Network

AU - WANG, Ran

AU - KWONG, Sam

AU - JIANG, Qingshan

AU - WONG, Ka-Chun

PY - 2016/1/12

Y1 - 2016/1/12

N2 - In this paper, we propose two stream-based active learning algorithms for single-hidden layer feed-forward neural networks (SLFNs) trained by extreme learning machine (ELM). Uncertainty and inconsistency are adopted as two sample selection criteria. Uncertainty reflects the nondeterminacy of a sample among different decision classes, which is calculated by information entropy or Gini-index. Inconsistency reflects the disagreement of the sample between its conditional features and decision labels, which is calculated by the lower approximations in fuzzy rough sets. Experimental results demonstrate that inconsistency-based strategy is more effective than uncertainty based strategy for SLFNs under stream-based environment.

AB - In this paper, we propose two stream-based active learning algorithms for single-hidden layer feed-forward neural networks (SLFNs) trained by extreme learning machine (ELM). Uncertainty and inconsistency are adopted as two sample selection criteria. Uncertainty reflects the nondeterminacy of a sample among different decision classes, which is calculated by information entropy or Gini-index. Inconsistency reflects the disagreement of the sample between its conditional features and decision labels, which is calculated by the lower approximations in fuzzy rough sets. Experimental results demonstrate that inconsistency-based strategy is more effective than uncertainty based strategy for SLFNs under stream-based environment.

KW - Active Learning

KW - Extreme Learning Machine

KW - Inconsistency

KW - SLFNs

KW - Uncertainty

UR - https://www.scopus.com/pages/publications/84964523738

U2 - 10.1109/SMC.2015.377

DO - 10.1109/SMC.2015.377

M3 - Conference paper (refereed)

SN - 9781479986965

SP - 2158

EP - 2163

BT - Proceedings : 2015 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2015

PB - IEEE

T2 - IEEE International Conference on Systems, Man, and Cybernetics, SMC 2015

Y2 - 9 October 2015 through 12 October 2015

ER -

Active Learning Based on Single-Hidden Layer Feed-Forward Neural Network

Abstract

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this