Computer Sciences Commons^™

Learning Internal State Memory Representations From Observation, Josiah Wong

Electronic Theses and Dissertations

Learning from Observation (LfO) is a machine learning paradigm that mimics how people learn in daily life: learning how to do something simply by watching someone else do it. LfO has been used in various applications, from video game agent creation to driving a car, but it has always been limited by the inability of an observer to know what a performing entity chooses to remember as they act in an environment. Various methods have either ignored the effects of memory or otherwise made simplistic assumptions about its structure. In this dissertation, we propose a new method, Memory Composition Learning, …

Optimization Algorithms For Deep Learning Based Medical Image Segmentations, Aliasghar Mortazi

Electronic Theses and Dissertations

Medical image segmentation is one of the fundamental processes to understand and assess the functionality of different organs and tissues as well as quantifying diseases and helping treatment planning. With ever increasing number of medical scans, the automated, accurate, and efficient medical image segmentation is as unmet need for improving healthcare. Recently, deep learning has emerged as one the most powerful methods for almost all image analysis tasks such as segmentation, detection, and classification and so in medical imaging. In this regard, this dissertation introduces new algorithms to perform medical image segmentation for different (a) imaging modalities, (b) number of …

Automated Synthesis Of Memristor Crossbar Networks, Dwaipayan Chakraborty

Electronic Theses and Dissertations

The advancement of semiconductor device technology over the past decades has enabled the design of increasingly complex electrical and computational machines. Electronic design automation (EDA) has played a significant role in the design and implementation of transistor-based machines. However, as transistors move closer toward their physical limits, the speed-up provided by Moore's law will grind to a halt. Once again, we find ourselves on the verge of a paradigm shift in the computational sciences as newer devices pave the way for novel approaches to computing. One of such devices is the memristor -- a resistor with non-volatile memory. Memristors can …

Utilizing Edge In Iot And Video Streaming Applications To Reduce Bottlenecks In Internet Traffic, Kutalmis Akpinar

Electronic Theses and Dissertations

There is a large increase in the surge of data over Internet due to the increasing demand on multimedia content. It is estimated that 80% of Internet traffic will be video by 2022, according to a recent study. At the same time, IoT devices on Internet will double the human population. While infrastructure standards on IoT are still nonexistent, enterprise solutions tend to encourage cloud-based solutions, causing an additional surge of data over the Internet. This study proposes solutions to bring video traffic and IoT computation back to the edges of the network, so that costly Internet infrastructure upgrades are …

A Study Of Perceptions On Incident Response Exercises, Information Sharing, Situational Awareness, And Incident Response Planning In Power Grid Utilities, Joseph Garmon

Electronic Theses and Dissertations

The power grid is facing increasing risks from a cybersecurity attack. Attacks that shut off electricity in Ukraine have already occurred, and successful compromises of the power grid that did not shut off electricity to customers have been privately disclosed in North America. The objective of this study is to identify how perceptions of various factors emphasized in the electric sector affect incident response planning. Methods used include a survey of 229 power grid personnel and the use of partial least squares structural equation modeling to identify causal relationships. This study reveals the relationships between perceptions by personnel responsible for …

On The Mental Workload Assessment Of Uplift Mapping Representations In Linked Data, Ademar Crotti Junior, Christophe Debruyne, Luca Longo, Declan O'Sullivan

Conference papers

Self-reporting procedures have been largely employed in literature to measure the mental workload experienced by users when executing a specific task. This research proposes the adoption of these mental workload assessment techniques to the task of creating uplift mappings in Linked Data. A user study has been performed to compare the mental workload of “manually” creating such mappings, using a formal mapping language and a text editor, to the use of a visual representation, based on the block metaphor, that generate these mappings. Two subjective mental workload instruments, namely the NASA Task Load Index and the Workload Profile, were applied …

The Use Of Deep Learning Distributed Representations In The Identification Of Abusive Text, Susan Mckeever, Hao Chen, Sarah Jane Delany

Conference papers

The selection of optimal feature representations is a critical step in the use of machine learning in text classification. Traditional features (e.g. bag of words and n-grams) have dominated for decades, but in the past five years, the use of learned distributed representations has become increasingly common. In this paper, we summarise and present a categorisation of the stateof-the-art distributed representation techniques, including word and sentence embedding models. We carry out an empirical analysis of the performance of the various feature representations using the scenario of detecting abusive comments. We compare classification accuracies across a range of off-the-shelf embedding models …

Optimal Separation And Strong Direct Sum For Randomized Query Complexity, E. Blais, Joshua Brody

Computer Science Faculty Works

We establish two results regarding the query complexity of bounded-error randomized algorithms. Bounded-error separation theorem. There exists a total function f : {0,1}^n -> {0,1} whose epsilon-error randomized query complexity satisfies overline{R}_epsilon(f) = Omega(R(f) * log 1/epsilon). Strong direct sum theorem. For every function f and every k >= 2, the randomized query complexity of computing k instances of f simultaneously satisfies overline{R}_epsilon(f^k) = Theta(k * overline{R}_{epsilon/k}(f)). As a consequence of our two main results, we obtain an optimal superlinear direct-sum-type theorem for randomized query complexity: there exists a function f for which R(f^k) = Theta(k log k * R(f)). …

Data-Driven Intelligent Scheduling For Long Running Workloads In Large-Scale Datacenters, Guoyao Xu

Wayne State University Dissertations

Cloud computing is becoming a fundamental facility of society today. Large-scale public or private cloud datacenters spreading millions of servers, as a warehouse-scale computer, are supporting most business of Fortune-500 companies and serving billions of users around the world. Unfortunately, modern industry-wide average datacenter utilization is as low as 6% to 12%. Low utilization not only negatively impacts operational and capital components of cost efficiency, but also becomes the scaling bottleneck due to the limits of electricity delivered by nearby utility. It is critical and challenge to improve multi-resource efficiency for global datacenters.

Additionally, with the great commercial success of …

Deep Learning Beyond Traditional Supervision, Shixing Chen

Wayne State University Dissertations

With the rapid development of innovative models and huge success on various applications, the field of deep learning has attracted enormous attention in computer vision, machine learning, and artificial intelligence. Countless researches have validated the superior performance and unprecedented extensiveness of deep learning models, especially with the advantages of high performance computing by GPUs and parallel computation. Nonetheless, drawbacks including strong dependency on supervision (sufficient labeled data) and monotonous usage of categorized labels are negatively interfering the advancement of deep learning.

In this dissertation, we plan to expose and exploit some possibilities of deep learning without using data and labels …

Evolution Of Portulacineae Marked By Gene Tree Conflict And Gene Family Expansion Associated With Adaptation To Harsh Environments, Ning Wang, Ya Yang, Michael J. Moore, Samuel F. Brockington, Joseph F. Walker, Joseph W. Brown, Bin Liang, Tao Feng, Caroline Edwards, Jessica Mikenas, Julia E. Olivieri, Vera Hutchison, Alfonso Timoneda, Tommy Stoughton, Raúl Puente, Lucas C. Majure, Urs Eggli, Stephen A. Smith

All Faculty Articles - School of Engineering and Computer Science

Several plant lineages have evolved adaptations that allow survival in extreme and harsh environments including many families within the plant clade Portulacineae (Caryophyllales) such as the Cactaceae, Didiereaceae, and Montiaceae. Here, using newly generated transcriptomic data, we reconstructed the phylogeny of Portulacineae and examined potential correlates between molecular evolution and adaptation to harsh environments. Our phylogenetic results were largely congruent with previous analyses, but we identified several early diverging nodes characterized by extensive gene tree conflict. For particularly contentious nodes, we present detailed information about the phylogenetic signal for alternative relationships. We also analyzed the frequency of gene duplications, confirmed …

Efficient Virtual Data Center Request Embedding Based On Row-Epitaxial And Batched Greedy Algorithms, Sivaranjani B, Surendran Doraiswamy

Turkish Journal of Electrical Engineering and Computer Sciences

Data centers are becoming the main backbone of and centralized repository for all cloud-accessible services in on-demand cloud computing environments. In particular, virtual data centers (VDCs) facilitate the virtualization of all data center resources such as computing, memory, storage, and networking equipment as a single unit. It is necessary to use the data center efficiently to improve its profitability. The essential factor that significantly influences efficiency is the average number of VDC requests serviced by the infrastructure provider, and the optimal allocation of requests improves the acceptance rate. In existing VDC request embedding algorithms, data center performance factors such as …

Hybrid Control Of Five-Phase Permanent Magnet Synchronous Machine Using Space Vector Modulation, Djamel Difi, Khaled Halbaoui, Djamel Boukhetala

Turkish Journal of Electrical Engineering and Computer Sciences

This paper aims to study the hybrid control of a five-phase permanent-magnet synchronous machine improved by the space vector modulation (SVM) technique. The torque ripples and currents will therefore be reduced. This control is based on the theory of hybrid dynamic systems (HDS), its discrete component is the voltage inverter which has a finite number of states controlling the continuous component that represents the machine. The results of the simulation made on MATLAB/Simulink are presented and discussed in order to check the performance of the strategy of the studied control. They show, in particular, the main advantages of this control …

Design Of A Portable And Low-Cost Mass-Sensitive Sensor With The Capability Of Measurements On Various Frequency Quartz Tuning Forks, Mehmet Altay Ünal, İsmai̇l Cengi̇z Koçum, Di̇lek Çökeli̇ler Serdaroğlu

Turkish Journal of Electrical Engineering and Computer Sciences

Recently, sensor and biosensor applications have become widespread and are now significant tools in the biomedical field and other areas. Since quartz tuning fork (QTF) resonance frequency depends on the mass adsorbed to its prongs, it is generally used to measure minor mass change and detect target analyte in picogram levels. This study is undertaken to design and fabricate a sensor device for the measurement of QTF transducers. When QTF sensor studies were investigated, it was found that explanations on the details of instrumentation part were limited, and in addition, there was no compact commercial products. In this study, a …

Generation Rescheduling Using Multiobjective Bilevel Optimization, Kiran Babu Vakkapatla, Srinivasa Varma Pinni

Turkish Journal of Electrical Engineering and Computer Sciences

This paper presents a new multiobjective optimization method that can be used for generation rescheduling in power systems. Generation rescheduling in restructured power systems is performed by the system operator for different operations like congestion management, day-ahead scheduling, and preventive maintenance. The nonlinear nature of the equations involved and the constraints on decision variables pose a challenge to find the global optimum. In order to find the global optimum using a genetic algorithm, a bilevel optimization method is proposed. In the proposed multiobjective optimization method, the objectives are classified as primary and secondary based on their relative importance. The best …

On The Output Regulation For Linear Fractional Systems, Jesus Alberto Meda Campana, Elba Cinthya Garcia Estrada, Julio Cesar Gomez Mancilla, Jose De Jesus Rubio Avila, Mario Ricardo Cruz Deviana, Ricardo Tapia Herrera

Turkish Journal of Electrical Engineering and Computer Sciences

In this work, the regulation problem is extended to the field of fractional-order linear systems considering the Caputo fractional derivative. The regulation equations are obtained on the basis of the Francis equations. It is also shown that the linear fractional regulator exists at $t=0$ only if the order of the plant is not greater than the order of the reference system.

Facial Re-Enactment, Speech Synthesis And The Rise Of The Deepfake, Nicholas Gardiner

Theses : Honours

Emergent technologies in the fields of audio speech synthesis and video facial manipulation have the potential to drastically impact our societal patterns of multimedia consumption. At a time when social media and internet culture is plagued by misinformation, propaganda and “fake news”, their latent misuse represents a possible looming threat to fragile systems of information sharing and social democratic discourse. It has thus become increasingly recognised in both academic and mainstream journalism that the ramifications of these tools must be examined to determine what they are and how their widespread availability can be managed.

This research project seeks to examine …

Efficient Local Comparison Of Images Using Krawtchouk Descriptors, Julian Deville

Online Theses and Dissertations

It is known that image comparison can prove cumbersome in both computational complexity and runtime, due to factors such as the rotation, scaling, and translation of the object in question. Due to the locality of Krawtchouk polynomials, relatively few descriptors are necessary to describe a given image, and this can be achieved with minimal memory usage. Using this method, not only can images be described efficiently as a whole, but specific regions of images can be described as well without cropping. Due to this property, queries can be found within a single large image, or collection of large images, which …

Image-Based Malware Classification: A Space Filling Curve Approach, Stephen O Shaughnessy

Conference Papers

Anti-virus (AV) software is effective at distinguishing between benign and malicious programs yet lack the ability to effectively classify malware into their respective family classes. AV vendors receive considerably large volumes of malicious programs daily and so classification is crucial to quickly identify variants of existing malware that would otherwise have to be manually examined. This paper proposes a novel method of visualizing and classifying malware using Space-Filling Curves (SFC's) in order to improve the limitations of AV tools. The classification models produced were evaluated on previously unseen samples and showed promising results, with precision, recall and accuracy scores of …

Predictive Modeling Of Webpage Aesthetics, Ang Chen

Masters Theses

"Aesthetics plays a key role in web design. However, most websites have been developed based on designers' inspirations or preferences. While perceptions of aesthetics are intuitive abilities of humankind, the underlying principles for assessing aesthetics are not well understood. In recent years, machine learning methods have shown promising results in image aesthetic assessment. In this research, we used machine learning methods to study and explore the underlying principles of webpage aesthetics"--Abstract, page iii.

Image-Based Roadway Assessment Using Convolutional Neural Networks, Weilian Song

Theses and Dissertations--Computer Science

Road crashes are one of the main causes of death in the United States. To reduce the number of accidents, roadway assessment programs take a proactive approach, collecting data and identifying high-risk roads before crashes occur. However, the cost of data acquisition and manual annotation has restricted the effect of these programs. In this thesis, we propose methods to automate the task of roadway safety assessment using deep learning. Specifically, we trained convolutional neural networks on publicly available roadway images to predict safety-related metrics: the star rating score and free-flow speed. Inference speeds for our methods are mere milliseconds, enabling …

Enhance Nmf-Based Recommendation Systems With Auxiliary Information Imputation, Fatemah Alghamedy

Theses and Dissertations--Computer Science

This dissertation studies the factors that negatively impact the accuracy of the collaborative filtering recommendation systems based on nonnegative matrix factorization (NMF). The keystone in the recommendation system is the rating that expresses the user's opinion about an item. One of the most significant issues in the recommendation systems is the lack of ratings. This issue is called "cold-start" issue, which appears clearly with New-Users who did not rate any item and New-Items, which did not receive any rating.

The traditional recommendation systems assume that users are independent and identically distributed and ignore the connections among users whereas the recommendation …

Optimal Gateway Placement In Low-Cost Smart Cities, Oluwashina Madamori

Theses and Dissertations--Computer Science

Rapid urbanization burdens city infrastructure and creates the need for local governments to maximize the usage of resources to serve its citizens. Smart city projects aim to alleviate the urbanization problem by deploying a vast amount of Internet-of-things (IoT) devices to monitor and manage environmental conditions and infrastructure. However, smart city projects can be extremely expensive to deploy and manage partly due to the cost of providing Internet connectivity via 5G or WiFi to IoT devices. This thesis proposes the use of delay tolerant networks (DTNs) as a backbone for smart city communication; enabling developing communities to become smart cities …

Healthcare Robotics: Key Factors That Impact Robot Adoption In Healthcare, Sujatha Alla, Pilar Pazos

Engineering Management & Systems Engineering Faculty Publications

In the current dynamic business environment, healthcare organizations are focused on improving patient satisfaction, performance, and efficiency. The healthcare industry is considered a complex system that is highly reliant of new technologies to support clinical as well as business processes. Robotics is one of such technologies that is considered to have the potential to increase efficiency in a wide range of clinical services. Although the use of robotics in healthcare is at the early stages of adoption, some studies have shown the capacity of this technology to improve precision, accessibility through less invasive procedures, and reduction of human error during …

Presenting A Product Design From Computer-Generated Imagery (Cgi), Mirjeta Mustafa

Theses and Dissertations

The purpose of this thesis is to determine a solution to overcome photography problems to present product design by using Computer Generated Imagery. CGI and virtual design now play a key role in the development of many materials, processes and products. Through this study has been explored, the impact of 3D render to exhibit the product design, as well as a brief history of photography and how it is used to present the product, the problems that have emerged during the use of photographs to present the product design have been discussed, and how new technologies have enabled us to …

Analizimi I Frameworks Për Krijimin E Frontend-It Të Një Aplikacioni, Marigona Ibrahimi

Theses and Dissertations

Ky hulumtim është fokusuar në analizimin e disa framework-ave të cilët përdoren për krijimin e frontend-eve të aplikacioneve. Framework-ët në ditët e sotme e kanë lehtësuar shumë zhvillimin e aplikacioneve, ata ofrojnë shumë klasa të gatshme, gjë që ia lehtësojnë punën zhvilluesve. Dita ditës numri i këtyre framework-ave po rritet, zhvilluesit duhet të zgjedhin njërin nga framework-ët për të punuar, por për t’a zgjedhur njërin nga ta, framework-ët duhet analizuar. Kërkimet për këtë studim janë bërë në web faqe përkatëse me literatura rreth framework-ave që janë caktuar për t’u analizuar, si dhe krahasimeve ndërmjet tyre. Qëllimi i këtij studimi është …

Përdorimi I Rrjetave Neurale Në Gërmimin E Të Dhënave, Abdylkadri Maksuti

Theses and Dissertations

Për një kohë të gjatë kompani apo biznese të ndryshme kanë mbledhur të dhëna të shumta të fushave të ndryshme. Shumë pak kompani apo biznese janë në dijeni për vlerën e këtyre të dhënave që mund të iu ofrojnë përfitime shumë të mëdha. Kështu që përmes teknologjive të ndryshme sic janë: gërmimi i të dhënave, rrjetat neurale ku kanë algoritme dhe teknika të ndryshme është arritur mundësia për ti nxjerrur këto të dhëna. Kështu që në ditët e sotme kanë marrur hov shumë të madh këto teknologji të cilat janë shumë të ndërlikuara.

Përmes kësaj teme diplome sqarohen se cka …

Krahasimi I Teknologjive Të Ruajtjes Se Informatave Online Si Onedrive, Google Drive Dhe Dropbox, Burim Bushi

Theses and Dissertations

Zhvillimi i këtyre teknologjive është vleresuar shumë sepse na mundëson ruajtjen e të dhënave online dhe qasjen në to në çfar do kohe që dëshirojm ne. Ka shumë lloje të tyre saqë konkurrenca ka bërë që secila të ofroj cilësi apo performanca më të mira. Zhvilluesit e këtyre teknologjive janë munduar që të punojnë, thjeshtësi në përdorim dhe efikasitet.

Në qoftë se kujtojmë kohen kur kjo teknologji nuk ka ekzistuar vërejm qe jemi perballur me veshtirësi të ndryshme ndërsa sot mjafton që të kyqemi në rrjet dhe të hapim një llogari personale pastaj qasja është edhe më e thjeshtë.

Është …

Siguria E Të Dhënave Në Cloud Computing, Ermal Bujupaj

Theses and Dissertations

Teknologjia e Cloud Computing gjatë viteve të fundit ka pasur një zhvillim të hovshëm, dhe me evoluimin e saj ka gjetur zbatim në gati shumicën e aplikacioneve apo platformave. Sot shumica e aplikacioneve si në web apo në platformat mobile mbështeten në infrastrukturat që ofrohen nga kompanitë e ndryshme në fushën e Cloud siç është AWS, Google Cloud Platform, Microsoft Azure etj.

Mirëpo me zhvillimin e Cloud si dhe me vendosjen e të dhënave në të, ngriten dy çështje mjaft të ndjeshme siç janë siguria dhe privatësia e të dhënave.

Siguria e të dhënave ka qenë vazhdimisht një çështje e …

Auxetic Regions In Large Deformations Of Periodic Frameworks, Ciprian S. Borcea, Ileana Streinu

Computer Science: Faculty Publications

In materials science, auxetic behavior refers to lateral widening upon stretching. We investigate the problem of finding domains of auxeticity in global deformation spaces of periodic frameworks. Case studies include planar periodic mechanisms constructed from quadrilaterals with diagonals as periods and other frameworks with two vertex orbits. We relate several geometric and kinematic descriptions.