MLP classifiers: Overtraining and solutions

Zheru Chi

MLP classifiers: Overtraining and solutions

Zheru Chi

The Hong Kong Polytechnic University

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

4 Citations (Scopus)

Abstract

Training a multi-layer perceptron (MLP) classifier is difficult to control and as a result its performance on unseen patterns is unpredicted. Overtraining is one of many problems in training an MLP classifier. In this paper, we first discuss the overtraining problem based on an artificial two-input two-category classification problem. We then suggest five solutions to the overtraining problem, which are supported by the experimental results.

Original language	English
Title of host publication	IEEE International Conference on Neural Networks - Conference Proceedings
Publisher	IEEE
Pages	2821-2824
Number of pages	4
Publication status	Published - 1 Dec 1995
Event	Proceedings of the 1995 IEEE International Conference on Neural Networks. Part 1 (of 6) - Perth, Australia Duration: 27 Nov 1995 → 1 Dec 1995

Conference

Conference	Proceedings of the 1995 IEEE International Conference on Neural Networks. Part 1 (of 6)
Country/Territory	Australia
City	Perth
Period	27/11/95 → 1/12/95

ASJC Scopus subject areas

Software

Cite this

@inproceedings{3d763a9ac300452ab0e234240e285375,

title = "MLP classifiers: Overtraining and solutions",

abstract = "Training a multi-layer perceptron (MLP) classifier is difficult to control and as a result its performance on unseen patterns is unpredicted. Overtraining is one of many problems in training an MLP classifier. In this paper, we first discuss the overtraining problem based on an artificial two-input two-category classification problem. We then suggest five solutions to the overtraining problem, which are supported by the experimental results.",

author = "Zheru Chi",

year = "1995",

month = dec,

day = "1",

language = "English",

pages = "2821--2824",

booktitle = "IEEE International Conference on Neural Networks - Conference Proceedings",

publisher = "IEEE",

address = "United States",

note = "Proceedings of the 1995 IEEE International Conference on Neural Networks. Part 1 (of 6) ; Conference date: 27-11-1995 Through 01-12-1995",

}

TY - GEN

T1 - MLP classifiers: Overtraining and solutions

AU - Chi, Zheru

PY - 1995/12/1

Y1 - 1995/12/1

N2 - Training a multi-layer perceptron (MLP) classifier is difficult to control and as a result its performance on unseen patterns is unpredicted. Overtraining is one of many problems in training an MLP classifier. In this paper, we first discuss the overtraining problem based on an artificial two-input two-category classification problem. We then suggest five solutions to the overtraining problem, which are supported by the experimental results.

AB - Training a multi-layer perceptron (MLP) classifier is difficult to control and as a result its performance on unseen patterns is unpredicted. Overtraining is one of many problems in training an MLP classifier. In this paper, we first discuss the overtraining problem based on an artificial two-input two-category classification problem. We then suggest five solutions to the overtraining problem, which are supported by the experimental results.

UR - http://www.scopus.com/inward/record.url?scp=0029457529&partnerID=8YFLogxK

M3 - Conference article published in proceeding or book

SP - 2821

EP - 2824

BT - IEEE International Conference on Neural Networks - Conference Proceedings

PB - IEEE

T2 - Proceedings of the 1995 IEEE International Conference on Neural Networks. Part 1 (of 6)

Y2 - 27 November 1995 through 1 December 1995

ER -

MLP classifiers: Overtraining and solutions

Abstract

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this