Application of a Scaled MNIST Dataset Blended with Natural Scene Background on ResNet

Alexander Marinov, Nhamo Mtetwa, Hadi Larijani

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Downloads (Pure)

Abstract

Deep learning (DL) has gained a lot of popularity in the science and business community. It has been successful in a range of applications, especially in computer vision. This paper presents results from applying scaled MNIST images dataset to a popular implementation of deep learning called ResNet. This is a valuable contribution because in general convolutional networks are not scale invariant. Our objective is to explore the behavior of a residual neural network when trained and evaluated using three different datasets of scaled MNIST images.
Original languageEnglish
Title of host publicationICBDE 2019
DOIs
Publication statusPublished - 31 Mar 2019

Fingerprint

Computer vision
Neural networks
Industry
Deep learning

Keywords

  • feature cnn
  • machine learning
  • scaling
  • mnist
  • resnet
  • deep learning

Cite this

@inproceedings{6b1e3de2bfdc420ba142d6403cb6e276,
title = "Application of a Scaled MNIST Dataset Blended with Natural Scene Background on ResNet",
abstract = "Deep learning (DL) has gained a lot of popularity in the science and business community. It has been successful in a range of applications, especially in computer vision. This paper presents results from applying scaled MNIST images dataset to a popular implementation of deep learning called ResNet. This is a valuable contribution because in general convolutional networks are not scale invariant. Our objective is to explore the behavior of a residual neural network when trained and evaluated using three different datasets of scaled MNIST images.",
keywords = "feature cnn, machine learning, scaling, mnist, resnet, deep learning",
author = "Alexander Marinov and Nhamo Mtetwa and Hadi Larijani",
note = "Acceptance in SAN DOI not found Accepted papers will be published in the International Conference Proceedings Series by ACM, which will be archived in the ACM Digital Library, and sent to be indexed by EI Compendex and Scopus and submitted to be reviewed by Thomson Reuters Conference Proceedings Citation Index (ISI Web of Science). AAM: made open in line with publisher policy (default ACM via Romeo) ET 5/7/19",
year = "2019",
month = "3",
day = "31",
doi = "10.1145/3322134.3322147",
language = "English",
isbn = "9781450361866",
booktitle = "ICBDE 2019",

}

Application of a Scaled MNIST Dataset Blended with Natural Scene Background on ResNet. / Marinov, Alexander; Mtetwa, Nhamo; Larijani, Hadi.

ICBDE 2019. 2019.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Application of a Scaled MNIST Dataset Blended with Natural Scene Background on ResNet

AU - Marinov, Alexander

AU - Mtetwa, Nhamo

AU - Larijani, Hadi

N1 - Acceptance in SAN DOI not found Accepted papers will be published in the International Conference Proceedings Series by ACM, which will be archived in the ACM Digital Library, and sent to be indexed by EI Compendex and Scopus and submitted to be reviewed by Thomson Reuters Conference Proceedings Citation Index (ISI Web of Science). AAM: made open in line with publisher policy (default ACM via Romeo) ET 5/7/19

PY - 2019/3/31

Y1 - 2019/3/31

N2 - Deep learning (DL) has gained a lot of popularity in the science and business community. It has been successful in a range of applications, especially in computer vision. This paper presents results from applying scaled MNIST images dataset to a popular implementation of deep learning called ResNet. This is a valuable contribution because in general convolutional networks are not scale invariant. Our objective is to explore the behavior of a residual neural network when trained and evaluated using three different datasets of scaled MNIST images.

AB - Deep learning (DL) has gained a lot of popularity in the science and business community. It has been successful in a range of applications, especially in computer vision. This paper presents results from applying scaled MNIST images dataset to a popular implementation of deep learning called ResNet. This is a valuable contribution because in general convolutional networks are not scale invariant. Our objective is to explore the behavior of a residual neural network when trained and evaluated using three different datasets of scaled MNIST images.

KW - feature cnn

KW - machine learning

KW - scaling

KW - mnist

KW - resnet

KW - deep learning

U2 - 10.1145/3322134.3322147

DO - 10.1145/3322134.3322147

M3 - Conference contribution

SN - 9781450361866

BT - ICBDE 2019

ER -