Analysis of uncertainty in different neural network structures using monte carlo dropout

This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science, 2023.

书目详细资料
Main Authors:	Islam, Md. Farhadul, Zabeen, Sarah, Bin Rahman, Fardin, Islam, Md. Azharul, Bin Kibria, Fahmid
其他作者:	Rasel, Annajiat Alim
格式:	Thesis
语言:	English
出版:	Brac University 2024
主题:	Deep learning Neural network Monte carlo dropout Uncertainty Neural network.
在线阅读:	http://hdl.handle.net/10361/22179

id	10361-22179
record_format	dspace
spelling	10361-221792024-01-17T21:02:41Z Analysis of uncertainty in different neural network structures using monte carlo dropout Islam, Md. Farhadul Zabeen, Sarah Bin Rahman, Fardin Islam, Md. Azharul Bin Kibria, Fahmid Rasel, Annajiat Alim Department of Computer Science and Engineering, Brac University Deep learning Neural network Monte carlo dropout Uncertainty Neural network. This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science, 2023. Cataloged from PDF version of thesis. Includes bibliographical references (pages 82-96). Deep learning technologies developed at an exponential rate throughout the years. Starting from Convolutional Neural Networks (CNNs) to Involutional Neural Net works (INNs), there are several neural network (NN) architectures today, including Vision Transformers (ViT), Graph Neural Networks (GNNs), Recurrent Neural Net works (RNNs) etc. However, uncertainty cannot be represented in these architec tures, which poses a significant difficulty for decision-making given that capturing the uncertainties of these state-of-the-art NN structures would aid in making spe cific judgments. Dropout is one method that may be implemented within Deep Learning (DL) networks as a technique to assess uncertainty. Dropout is applied at the inference phase to measure the uncertainty of these neural network models. This approach, commonly known as Monte Carlo Dropout (MCD), works well as a low-complexity estimation to compute uncertainty. MCD is a widely used approach to measure uncertainty in DL models, but majority of the earlier works focus on only a particular application. Furthermore, there are many state-of-the-art (SOTA) NNs that remain unexplored, with regards to that of uncertainty evaluation. There fore an up-to-date roadmap and benchmark is required in this field of study. Our study revolved around a comprehensive analysis of the MCD approach for assessing model uncertainty in neural network models with a variety of datasets. Besides, we include SOTA NNs to explore the untouched models regarding uncertainty. In addition, we demonstrate how the model may perform better with less uncertainty by modifying NN topologies, which also reveals the causes of a model’s uncertainty. Using the results of our experiments and subsequent enhancements, we also discuss the various advantages and costs of using MCD in these NN designs. While working with reliable and robust models we propose two novel architectures, which provide outstanding performances in medical image diagnosis. Md. Farhadul Islam Sarah Zabeen Fardin Bin Rahman Md. Azharul Islam Fahmid Bin Kibria B.Sc. in Computer Science 2024-01-17T08:21:11Z 2024-01-17T08:21:11Z 2023 2023-01 Thesis ID: 22341042 ID: 19241004 ID: 20101592 ID: 19301257 ID: 19201063 http://hdl.handle.net/10361/22179 en Brac University theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. 96 pages application/pdf Brac University
institution	Brac University
collection	Institutional Repository
language	English
topic	Deep learning Neural network Monte carlo dropout Uncertainty Neural network.
spellingShingle	Deep learning Neural network Monte carlo dropout Uncertainty Neural network. Islam, Md. Farhadul Zabeen, Sarah Bin Rahman, Fardin Islam, Md. Azharul Bin Kibria, Fahmid Analysis of uncertainty in different neural network structures using monte carlo dropout
description	This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science, 2023.
author2	Rasel, Annajiat Alim
author_facet	Rasel, Annajiat Alim Islam, Md. Farhadul Zabeen, Sarah Bin Rahman, Fardin Islam, Md. Azharul Bin Kibria, Fahmid
format	Thesis
author	Islam, Md. Farhadul Zabeen, Sarah Bin Rahman, Fardin Islam, Md. Azharul Bin Kibria, Fahmid
author_sort	Islam, Md. Farhadul
title	Analysis of uncertainty in different neural network structures using monte carlo dropout
title_short	Analysis of uncertainty in different neural network structures using monte carlo dropout
title_full	Analysis of uncertainty in different neural network structures using monte carlo dropout
title_fullStr	Analysis of uncertainty in different neural network structures using monte carlo dropout
title_full_unstemmed	Analysis of uncertainty in different neural network structures using monte carlo dropout
title_sort	analysis of uncertainty in different neural network structures using monte carlo dropout
publisher	Brac University
publishDate	2024
url	http://hdl.handle.net/10361/22179
work_keys_str_mv	AT islammdfarhadul analysisofuncertaintyindifferentneuralnetworkstructuresusingmontecarlodropout AT zabeensarah analysisofuncertaintyindifferentneuralnetworkstructuresusingmontecarlodropout AT binrahmanfardin analysisofuncertaintyindifferentneuralnetworkstructuresusingmontecarlodropout AT islammdazharul analysisofuncertaintyindifferentneuralnetworkstructuresusingmontecarlodropout AT binkibriafahmid analysisofuncertaintyindifferentneuralnetworkstructuresusingmontecarlodropout
_version_	1814309728721829888

Analysis of uncertainty in different neural network structures using monte carlo dropout

相似书籍