Hyperparameter Optimization for Deep Learning-based Automatic Melanoma Diagnosis System

IF 0.8 Q4 ENGINEERING, BIOMEDICAL
T. Nagaoka
{"title":"Hyperparameter Optimization for Deep Learning-based Automatic Melanoma Diagnosis System","authors":"T. Nagaoka","doi":"10.14326/abe.9.225","DOIUrl":null,"url":null,"abstract":"Deep learning is widely used in the development of automatic diagnosis systems for melanoma. However, there are some parameters called hyperparameters which should be set arbitrarily. Optimum setting of hyperparameters is challenging. The dermoscopic images on the database are trained on GoogLeNet. The hyperparameters verified in this study were random seed, solver type, base learning rate, epoch, and batch size. By using a genetic algorithm, these hyperparameters were optimized to obtain higher validation accuracy than other methods such as brute force or Bayesian optimization. The highest validation accuracy was 89.75%. The best hyperparameter settings were: 2 for random seed, RMSProp for solver type, 0.0001 for base learning rate, 30 for epoch, 32 for batch size, and 368 seconds for training time. Using the genetic algorithm, we successful-ly set the hyperparameters for efficient deep learning. Using the system developed in this study, we plan to search for a broader range of hyperparameters and identify multiple groups including lesions other than melanoma.","PeriodicalId":54017,"journal":{"name":"Advanced Biomedical Engineering","volume":"1 1","pages":""},"PeriodicalIF":0.8000,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"4","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Advanced Biomedical Engineering","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.14326/abe.9.225","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"ENGINEERING, BIOMEDICAL","Score":null,"Total":0}
引用次数: 4

Abstract

Deep learning is widely used in the development of automatic diagnosis systems for melanoma. However, there are some parameters called hyperparameters which should be set arbitrarily. Optimum setting of hyperparameters is challenging. The dermoscopic images on the database are trained on GoogLeNet. The hyperparameters verified in this study were random seed, solver type, base learning rate, epoch, and batch size. By using a genetic algorithm, these hyperparameters were optimized to obtain higher validation accuracy than other methods such as brute force or Bayesian optimization. The highest validation accuracy was 89.75%. The best hyperparameter settings were: 2 for random seed, RMSProp for solver type, 0.0001 for base learning rate, 30 for epoch, 32 for batch size, and 368 seconds for training time. Using the genetic algorithm, we successful-ly set the hyperparameters for efficient deep learning. Using the system developed in this study, we plan to search for a broader range of hyperparameters and identify multiple groups including lesions other than melanoma.
基于深度学习的黑色素瘤自动诊断系统的超参数优化
深度学习被广泛应用于黑色素瘤自动诊断系统的开发。然而,有一些被称为超参数的参数可以任意设置。超参数的最佳设置是具有挑战性的。数据库中的皮肤镜图像在GoogLeNet上进行训练。本研究验证的超参数包括随机种子、求解器类型、基本学习率、epoch和批大小。通过使用遗传算法对这些超参数进行优化,以获得比其他方法(如蛮力或贝叶斯优化)更高的验证精度。最高验证准确率为89.75%。最佳超参数设置为:随机种子为2,求解器类型为RMSProp,基本学习率为0.0001,epoch为30,批大小为32,训练时间为368秒。利用遗传算法,我们成功地设置了高效深度学习的超参数。使用本研究中开发的系统,我们计划搜索更大范围的超参数,并识别包括黑色素瘤以外病变的多个组。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Advanced Biomedical Engineering
Advanced Biomedical Engineering ENGINEERING, BIOMEDICAL-
CiteScore
1.40
自引率
10.00%
发文量
15
审稿时长
15 weeks
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信