Hyper parameters selection for image classification in convolutional neural networks

Sajid Nazir; Shushma Patel; Dilip Patel

doi:10.1109/ICCI-CC.2018.8482081

Hyper parameters selection for image classification in convolutional neural networks

Sajid Nazir, Shushma Patel, Dilip Patel

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

7 Citations (Scopus)

324 Downloads (Pure)

Abstract

Cognitive image processing is made possible by the availability of faster and cheaper memories, increased processing power of multicore processors and the explosive growth in visual data generation. Object classification remains an active research area with rapid advancements. The recent advances in deep neural networks are providing better than expected results for
image classification. The process involves training models with large labelled datasets to learn the underlying features from various image classes to support cognitive inferences on the test data. By training on large annotated image datasets contextual and semantic information can be automatically extracted from the images. An important determinant of better results is the choice of hyper parameters which is difficult to get right. In this paper we empirically investigate the effect of selected hyper parameters of a convolutional neural network on CIFAR-10 dataset and provide results to demonstrate their effect and
importance for image classification.

Original language	English
Title of host publication	17th IEEE International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC 2018)
Publisher	IEEE
Pages	401-407
Number of pages	7
ISBN (Electronic)	978-1-5386-3360-1
ISBN (Print)	9781538633601
DOIs	https://doi.org/10.1109/ICCI-CC.2018.8482081
Publication status	Published - 8 Oct 2018
Event	2018 IEEE 17th International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC) - Berkeley, United States Duration: 16 Jul 2018 → 18 Jul 2018 Conference number: 17 https://ieeexplore.ieee.org/servlet/opac?punumber=8470026

Conference

Conference	2018 IEEE 17th International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC)
Country/Territory	United States
City	Berkeley
Period	16/07/18 → 18/07/18
Internet address	https://ieeexplore.ieee.org/servlet/opac?punumber=8470026

Keywords

Cognitive image processing
Convolution
Deep learning
Hyper parameters
Neural networks

ASJC Scopus subject areas

Artificial Intelligence
Information Systems
Cognitive Neuroscience

Documents and Links

10.1109/ICCI-CC.2018.8482081

Nazir, S. et al (2018) Hyper parameters selection for image classification in convolutional neural networks
© 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Author accepted manuscript, 698 KB

http://www.ucalgary.ca/icci_cc/iccicc-18/call-papers

Cite this

@inproceedings{951a1a2875db4ce4b14e0c8f9875b627,

title = "Hyper parameters selection for image classification in convolutional neural networks",

abstract = "Cognitive image processing is made possible by the availability of faster and cheaper memories, increased processing power of multicore processors and the explosive growth in visual data generation. Object classification remains an active research area with rapid advancements. The recent advances in deep neural networks are providing better than expected results forimage classification. The process involves training models with large labelled datasets to learn the underlying features from various image classes to support cognitive inferences on the test data. By training on large annotated image datasets contextual and semantic information can be automatically extracted from the images. An important determinant of better results is the choice of hyper parameters which is difficult to get right. In this paper we empirically investigate the effect of selected hyper parameters of a convolutional neural network on CIFAR-10 dataset and provide results to demonstrate their effect andimportance for image classification. ",

keywords = "Cognitive image processing, Convolution, Deep learning, Hyper parameters, Neural networks",

author = "Sajid Nazir and Shushma Patel and Dilip Patel",

note = "Acceptance in SAN; 2018 IEEE 17th International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC) ; Conference date: 16-07-2018 Through 18-07-2018",

year = "2018",

month = oct,

day = "8",

doi = "10.1109/ICCI-CC.2018.8482081",

language = "English",

isbn = "9781538633601",

pages = "401--407",

booktitle = "17th IEEE International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC 2018)",

publisher = "IEEE",

address = "United States",

url = "https://ieeexplore.ieee.org/servlet/opac?punumber=8470026",

}

Nazir, S, Patel, S & Patel, D 2018, Hyper parameters selection for image classification in convolutional neural networks. in 17th IEEE International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC 2018). IEEE, pp. 401-407, 2018 IEEE 17th International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC), Berkeley, California, United States, 16/07/18. https://doi.org/10.1109/ICCI-CC.2018.8482081

TY - GEN

T1 - Hyper parameters selection for image classification in convolutional neural networks

AU - Nazir, Sajid

AU - Patel, Shushma

AU - Patel, Dilip

N1 - Conference code: 17

PY - 2018/10/8

Y1 - 2018/10/8

N2 - Cognitive image processing is made possible by the availability of faster and cheaper memories, increased processing power of multicore processors and the explosive growth in visual data generation. Object classification remains an active research area with rapid advancements. The recent advances in deep neural networks are providing better than expected results forimage classification. The process involves training models with large labelled datasets to learn the underlying features from various image classes to support cognitive inferences on the test data. By training on large annotated image datasets contextual and semantic information can be automatically extracted from the images. An important determinant of better results is the choice of hyper parameters which is difficult to get right. In this paper we empirically investigate the effect of selected hyper parameters of a convolutional neural network on CIFAR-10 dataset and provide results to demonstrate their effect andimportance for image classification.

AB - Cognitive image processing is made possible by the availability of faster and cheaper memories, increased processing power of multicore processors and the explosive growth in visual data generation. Object classification remains an active research area with rapid advancements. The recent advances in deep neural networks are providing better than expected results forimage classification. The process involves training models with large labelled datasets to learn the underlying features from various image classes to support cognitive inferences on the test data. By training on large annotated image datasets contextual and semantic information can be automatically extracted from the images. An important determinant of better results is the choice of hyper parameters which is difficult to get right. In this paper we empirically investigate the effect of selected hyper parameters of a convolutional neural network on CIFAR-10 dataset and provide results to demonstrate their effect andimportance for image classification.

KW - Cognitive image processing

KW - Convolution

KW - Deep learning

KW - Hyper parameters

KW - Neural networks

U2 - 10.1109/ICCI-CC.2018.8482081

DO - 10.1109/ICCI-CC.2018.8482081

M3 - Conference contribution

SN - 9781538633601

SP - 401

EP - 407

BT - 17th IEEE International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC 2018)

PB - IEEE

T2 - 2018 IEEE 17th International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC)

Y2 - 16 July 2018 through 18 July 2018

ER -

Hyper parameters selection for image classification in convolutional neural networks

Abstract

Conference

Keywords

ASJC Scopus subject areas

Documents and Links

Fingerprint

Cite this