Deep Evolution of Feature Representations for Handwritten Digit Recognition

Files in This Item:
File Description SizeFormat 
agapitos2015a.pdf465.95 kBAdobe PDFDownload
Title: Deep Evolution of Feature Representations for Handwritten Digit Recognition
Authors: Agapitos, AlexandrosO'Neill, MichaelNicolau, MiguelFagan, DavidKattan, AhmedCurran, Kathleen M.
Permanent link: http://hdl.handle.net/10197/8274
Date: 28-May-2015
Online since: 2017-01-16T18:01:25Z
Abstract: A training protocol for learning deep neural networks, called greedy layer-wise training, is applied to the evolution of a hierarchical, feed-forward Genetic Programming based system for feature construction and object recognition. Results on a popular handwritten digit recognition benchmark clearly demonstrate that two layers of feature transformations improves generalisation compared to a single layer. In addition, we show that the proposed system outperforms several standard Genetic Programming systems, which are based on hand-designed features, and use different program representations and fitness functions.
Type of material: Conference Publication
Publisher: IEEE
Copyright (published version): 2015 IEEE
Keywords: Object recognitionFeature extractionGenetic algorithmsHandwritten character recognition
DOI: 10.1109/CEC.2015.7257189
Language: en
Status of Item: Peer reviewed
Conference Details: 2015 IEEE Congress on Evolutionary Computation (CEC), Sendai, Japan, 25-28 May 2015
This item is made available under a Creative Commons License: https://creativecommons.org/licenses/by-nc-nd/3.0/ie/
Appears in Collections:Business Research Collection
CASL Research Collection

Show full item record

SCOPUSTM   
Citations 50

6
Last Week
0
Last month
checked on Sep 11, 2020

Page view(s) 50

1,322
Last Week
5
Last month
17
checked on Apr 17, 2021

Download(s)

110
checked on Apr 17, 2021

Google ScholarTM

Check

Altmetric


If you are a publisher or author and have copyright concerns for any item, please email research.repository@ucd.ie and the item will be withdrawn immediately. The author or person responsible for depositing the article will be contacted within one business day.