Computer Engineering Commons^™

Teachability And Interpretability In Reinforcement Learning, Jeevan Rajagopal

School of Computing: Dissertations, Theses, and Student Research

There have been many recent advancements in the field of reinforcement learning, starting from the Deep Q Network playing various Atari 2600 games all the way to Google Deempind's Alphastar playing competitively in the game StarCraft. However, as the field challenges more complex environments, the current methods of training models and understanding their decision making become less effective. Currently, the problem is partially dealt with by simply adding more resources, but the need for a better solution remains.

This thesis proposes a reinforcement learning framework where a teacher or entity with domain knowledge of the task to complete can assist …

“The Revolution Will Not Be Supervised": An Investigation Of The Efficacy And Reasoning Process Of Self-Supervised Representations, Atharva Tendle

School of Computing: Dissertations, Theses, and Student Research

Transfer learning technique enables training Deep Learning (DL) models in a data-efficient way for solving computer vision tasks. It involves pretraining a DL model to learn representations from a large and general-purpose source dataset, then fine-tuning the model using the task-specific target dataset. The dominant supervised learning (SL) approach for pretraining representations suffers from some limitations that include expensive labeling and poor generalizability. Recent advancements in the self-supervised learning (SSL) approach made it possible to learn effective representations from unlabeled data. The performance of the fine-tuned DL models based on pretrained SSL representations is on par with the state-of-the-art pretrained …

A Fully-Automated, Deep Learning-Based Framework For Ct-Based Localization, Segmentation, Verification And Planning Of Metastatic Vertebrae, Tucker Netherton, Tucker James Netherton

Dissertations and Theses (Open Access)

Palliative radiotherapy is an effective treatment for the palliation of symptoms caused by vertebral metastases. Visible evidence of disease is localized on medical images as part of the treatment planning process. However, complicating factors such as time pressures, anatomic variants in the spine, and similarities in adjacent vertebrae are associated with wrong level treatments of the spine. In addition, erroneous manual contouring of anatomic structures is a major failure mode in radiotherapy treatment planning.

The purpose of this study is to mitigate the challenges associated with treatment planning of the spine by automating the treatment planning process for three-dimensional conformal …

Improving Treatment Of Local Liver Ablation Therapy With Deep Learning And Biomechanical Modeling, Brian Anderson, Kristy Brock, Laurence Court, Carlos Eduardo Cardenas, Erik Cressman, Ankit Patel

Dissertations and Theses (Open Access)

In the United States, colorectal cancer is the third most diagnosed cancer, and 60-70% of patients will develop liver metastasis. While surgical liver resection of metastasis is the standard of care for treatment with curative intent, it is only avai lable to about 20% of patients. For patients who are not surgical candidates, local percutaneous ablation therapy (PTA) has been shown to have a similar 5-year overall survival rate. However, PTA can be a challenging procedure, largely due to spatial uncertainties in the localization of the ablation probe, and in measuring the delivered ablation margin.

For this work, we hypothesized …

A Golden Age For Computing Frontiers, A Dark Age For Computing Education?, Christof Teuscher

Electrical and Computer Engineering Faculty Publications and Presentations

There is no doubt that the body of knowledge spanned by the computing disciplines has gone through an unprecedented expansion, both in depth and breadth, over the last century. In this position paper, we argue that this expansion has led to a crisis in computing education: quite literally the vast majority of the topics of interest of this conference are not taught at the undergraduate level and most graduate courses will only scratch the surface of a few selected topics. But alas, industry is increasingly expecting students to be familiar with emerging topics, such as neuromorphic, probabilistic, and quantum computing, …

Low-Power And Reconfigurable Asynchronous Asic Design Implementing Recurrent Neural Networks, Spencer Nelson

Graduate Theses and Dissertations

Artificial intelligence (AI) has experienced a tremendous surge in recent years, resulting in high demand for a wide array of implementations of algorithms in the field. With the rise of Internet-of-Things devices, the need for artificial intelligence algorithms implemented in hardware with tight design restrictions has become even more prevalent. In terms of low power and area, ASIC implementations have the best case. However, these implementations suffer from high non-recurring engineering costs, long time-to-market, and a complete lack of flexibility, which significantly hurts their appeal in an environment where time-to-market is so critical. The time-to-market gap can be shortened through …

Interactive Search Vs. Automatic Search: An Extensive Study On Video Retrieval, Phuong-Anh Nguyen, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

This article conducts user evaluation to study the performance difference between interactive and automatic search. Particularly, the study aims to provide empirical insights of how the performance landscape of video search changes, with tens of thousands of concept detectors freely available to exploit for query formulation. We compare three types of search modes: free-to-play (i.e., search from scratch), non-free-to-play (i.e., search by inspecting results provided by automatic search), and automatic search including concept-free and concept-based retrieval paradigms. The study involves a total of 40 participants; each performs interactive search over 15 queries of various difficulty levels using two search modes …

A Hyperelastic Porous Media Framework For Ionic Polymer-Metal Composites And Characterization Of Transduction Phenomena Via Dimensional Analysis And Nonlinear Regression, Zakai J. Olsen

UNLV Theses, Dissertations, Professional Papers, and Capstones

Ionic polymer-metal composites (IPMC) are smart materials that exhibit large deformation in response to small applied voltages, and conversely generate detectable electrical signals in response to mechanical deformations. The study of IPMC materials is a rich field of research, and an interesting intersection of material science, electrochemistry, continuum mechanics, and thermodynamics. Due to their electromechanical and mechanoelectrical transduction capabilities, IPMCs find many applications in robotics, soft robotics, artificial muscles, and biomimetics. This study aims to investigate the dominating physical phenomena that underly the actuation and sensing behavior of IPMC materials. This analysis is made possible by developing a new, hyperelastic …

Ship-Gan: Generative Modeling Based Maritime Traffic Simulator, Chaithanya Shankaramurthy Basrur, Arambam James Singh, Arunesh Sinha, Akshat Kumar

Research Collection School Of Computing and Information Systems

Modeling vessel movement in a maritime environment is an extremely challenging task given the complex nature of vessel behavior. Several existing multiagent maritime decision making frameworks require access to an accurate traffic simulator. We develop a system using electronic navigation charts to generate realistic and high fidelity vessel traffic data using Generative Adversarial Networks (GANs). Our proposed Ship-GAN uses a conditional Wasserstein GAN to model a vessel’s behavior. The generator can simulate the travel time of vessels across different maritime zones conditioned on vessels’ speeds and traffic intensity. Furthermore, it can be used as an accurate simulator for prior decision …

Securing Fog Federation From Behavior Of Rogue Nodes, Mohammed Saleh H. Alshehri

Graduate Theses and Dissertations

As the technological revolution advanced information security evolved with an increased need for confidential data protection on the internet. Individuals and organizations typically prefer outsourcing their confidential data to the cloud for processing and storage. As promising as the cloud computing paradigm is, it creates challenges; everything from data security to time latency issues with data computation and delivery to end-users. In response to these challenges CISCO introduced the fog computing paradigm in 2012. The intent was to overcome issues such as time latency and communication overhead and to bring computing and storage resources close to the ground and the …

A Deep Analysis And Algorithmic Approach To Solving Complex Fitness Issues In Collegiate Student Athletes, Holly N. Puckett

Honors College Theses

Sports are not simply an entertainment source. For many, it creates a sense of community, support, and trust among both fans and athletes alike. In order to continue the sense of community sports provides, athletes must be properly cared for in order to perform at the highest level possible. Thus, their fitness and health must be monitored continuously. In a professional sense, one can expect individualized attention to athletes daily due to an abundance of funding and resources. However, when looking at college communities and student athletes within them, the number of athletes per athletic trainer increases due to both …

Mapping Renewal: How An Unexpected Interdisciplinary Collaboration Transformed A Digital Humanities Project, Elise Tanner, Geoffrey Joseph

Digital Initiatives Symposium

Funded by a National Endowment for Humanities (NEH) Humanities Collections and Reference Resources Foundations Grant, the UA Little Rock Center for Arkansas History and Culture’s “Mapping Renewal” pilot project focused on creating access to and providing spatial context to archival materials related to racial segregation and urban renewal in the city of Little Rock, Arkansas, from 1954-1989. An unplanned interdisciplinary collaboration with the UA Little Rock Arkansas Economic Development Institute (AEDI) has proven to be an invaluable partnership. One team member from each department will demonstrate the Mapping Renewal website and discuss how the collaborative process has changed and shaped …

Student Academic Conference, Caitlin Brooks

Student Academic Conference

No abstract provided.

Mesoscopic Traffic Simulation Model And Calibration Considering Stretching-Segment Design, Zhaocheng He, Xuanhua Lin, Peilin Nie, Ronghui Zhang

Journal of System Simulation

Abstract: In order to make the simulation model fit the characteristics of urban traffic, both high accuracy and high performance, a lightweight mesoscopic traffic simulation system and the process of calibration are established. The speed-density model and vertical queue model are equivalent to the vehicle movement processes, the simulation accuracy and calibration efficiency are improved by the stretching-segment design at urban intersections of vertical queuing model, and the real individual vehicle information is used as the calibration data source. The application of the model in Xuancheng urban road network shows that, compared with the vertical queuing model, it can …

Modeling And Simulation Of Electric Vehicle Industry Development Based On System Dynamics, Yueqiang Fu, Tiantian Xia

Journal of System Simulation

Abstract: New energy electric vehicle are the main trend of automobile industry upgrading. It plays an important role in ensuring the energy security and improving the ecological environment. It is of theoretical and practical significance to carry out research on the development of new energy electric vehicles. The affecting factors are systematically analyzed, the causal relationship model and stock flow model are established, and the dynamic equation of the model are determined and the parameter assignments are made. The model is verified and the system simulation and analysis are performed. The development trend and main influencing factors of …

Small Fault Detection Based On Cumulative Sum Of Neighbor Statistic, Xiaoping Guo, Jiajun Gao, Jianbin Guo, Li Yuan

Journal of System Simulation

Abstract: Aiming at the small faults and the common data non-linear problems of industrial process, a fault detection method based oncumulative sum of neighbor statistic (CUSUM-NS) is proposed. Mutual information principal component analysis (MIPCA) is used to reduce the dimension of training data, and the principal components based on mutual information are extracted to construct a new sample space. For the new sample space after dimensionality reduction, the nonlinear features of the process data can be fully extracted through the distance square sum statistics of k nearest neighbors. Cumulative summation(CUSUM) method is used to accumulate the sum of squares of …

Visual Simulation Platform For Visible Light Reconnaissance Load Of Unmanned Aerial Vehicle, Yuzhou Chen, Li Yuan, Qinglin Wang, Zhang Qing, Jinyuan Zhang

Journal of System Simulation

Abstract: In view of the simulation and evaluation requirement of the visual system parameters on the performance of video imaging during the operation and reconnaissance of unmanned aerial vehicle, a visual simulation platform for the reconnaissance load is designed and constructed. The collected video is processed according to the visual system parameters and the flight parameters, and the support for the evaluation and the index design of the unmanned aerial vehicle reconnaissance load system is provided, and the guidance is provided for the flight parameters and the flight track setting when the unmanned aerial vehicle reconnaissance and operation …

Research On Some Questions Of Simulation Body Of Knowledge, Xiaogang Qiu, Duan Hong, Xie Xu, Bin Chen

Journal of System Simulation

Abstract: Simulation body of knowledge (BOK) is the knowledge required to conduct Modeling and Simulation (M&S) activities, which is a logic system consisting of concepts, propositions, and inferences that are tightly related to each other. The simulation BOK organizes the M&S knowledge in a hierarchical way, reflects the composition and structure of the knowledge in the M&S domain. The establishment of the simulation BOK is crucial to advance the M&S research and education. The requirements for establishing the simulation BOK are summarized, three basic features of the simulation knowledge, practical, systematical, and epochal are discussed, the challenges of establishing the …

Study On Composition Of Simulation Discipline Knowledge Areas, Duan Hong, Xiaogang Qiu, Xie Xu, Rusheng Ju

Journal of System Simulation

Abstract: Many disciplines, such as Software Engineering, Automation, have sorted out and formed their own knowledge areas and constructed their Body of Knowledge to steer teaching and study efforts. Describing the composition of Modeling and Simulation body of knowledge from the perspective of knowledge area plays an important role in the simulation engineering education and the development of simulation technology. According to the needs of simulation teaching, the body of knowledge of simulation discipline is divided into three levels, knowledge area, knowledge unit and knowledge topic. On the basis of reviewing the current status of the research and the role …

Research On Combat Simulation Body Of Knowledge, Xie Xu, Xiaogang Qiu, Duan Hong, Kedi Huang

Journal of System Simulation

Abstract: Combat simulation is an important research method in modern military domain, since it provides a virtual battlespace for entities of various types that are involved in a battle to interact with each other. Over last several decades, combat simulation has been widely applied in different applications, and as a result the body of knowledge of combat simulation has been extended a lot. The relevant articles and textbooks in combat simulation are extensively investigated, and a three-layer structure to organize the body of knowledge of combat simulation is proposed. Eleven knowledge areas that should be included in the combat simulation …

Personalized Game Recommendation Method Based On Implicit Feedback, Sha Jing, Gongli Zeng, Yang Yang, Wei Yao

Journal of System Simulation

Abstract: Traditional recommendation systems often use explicit feedback for personalized recommendations. But the explicit feedback data is not easy to obtain, and the quality is poor, and the recommendation results unable to meet the requitrment of the user. Implicit feedback data is easier to obtain and can provide users with the better content. A personalized game recommendation method based on implicit feedback data is proposed. The method builds an implicit feedback recommendation model for game user data based on implicit feedback data such as the game duration and game numbers. A personalized recommendation of the game is implemented through an …

Mesh Solid Construction Algorithm Of Spiral Bevel Gear Based On Virtual Collision Body, Cheng'en Li, Xiangjun Zou, Zeqin Zeng, Jianhua He, Li Hui, Zhaofeng Huang

Journal of System Simulation

Abstract: In order to improve the production automation, intelligence level and production efficiency of the tractor rear axle, the spiral bevel gear mesh entity based on the virtual collision body is constructed to carry out the human-machine interaction virtual simulation experiment of the tractor rear axle. The spiral bevel gear made by Gleason as is taken an example, the processing technology of arc spur gear is analyzed, the kinematics is used to establish a mathematical model of the spiral bevel gear forming process. The differential and interpolation methods are used to fit the gear curve and surface, the contour point …

Gas-Liquid Two-Phase Flow Pattern Recognition Method Based On Convolutional Neural Network, Weiguo Tong, Xuechun Pang, Genghong Zhu

Journal of System Simulation

Abstract: Aiming at the low recognition rate and subjectivity in two-phase flow pattern recognition, a method based on Landweber iterative image reconstruction algorithm and convolutional neural network is proposed. Landweber iterative image reconstruction algorithm is used to obtain the flow pattern images and build the flow pattern image database. By means of the flow pattern identification on, different convolution layers in VGG16 network and different size and resolution of the data set samples, the parameters of network frozen convolutional layer and input image are determined.The experimental results show that the combined method of resistance tomography and convolutional neural network …

Operation Resilience Optimization Of Power System Considering Generalized Energy Storage, Weiqing Sun, Zhang Jie, Ye Lei, Han Dong

Journal of System Simulation

Abstract: Based on the definition and principle of power system resilience, combining the traditional energy storage equipment with the demand response, an optimization method of power system operation resilience considering generalized energy storage is proposed. Five attack schemes based on topology are evaluated according to four indexes, and the most disadvantageous attack strategy is selected to simulate the damage of power system. Taking the minimum operating cost of the system as the objective, a combined scheduling model involving wind power station and energy storage unit is carried out. A generalized energy storage scheduling method is proposed to improve the …

Study On Relay Selection Algorithm Based On Swipt Wireless Cooperative Network, Qun Fang, Xukai Chen, He Xin, Yujun Zhu, Yiyang Liu, Yangyang Fang, Heju Li

Journal of System Simulation

Abstract: Although the wireless cooperative networks (WCN) significantly improve the communication quality and network throughput of wireless communications. However, it still faces the challenge of the network lifetime and the energy replenishment due to the energy limitation of relay nodes. In order to tackle this challenge, A multi-relay wireless cooperative network combined with the simultaneous wireless information and power transfer (SWIPT) is proposed and the theoretical performance under the Nakagami-m assumption is analyzed. The selection of the optimal relay node for wireless cooperative networks in the framework of the proposed system is studied. The outage probability of the system …

Compensation Sliding Cross Coupling Control Research Of Cartesian Coordinate Robot, Wang Wei, Zhimei Chen, Zhenyan Wang

Journal of System Simulation

Abstract: For a typical Cartesian coordinate robot controls precision is low, based on a single-axis mathematical model, a contour error model for a typical robot whose axes are orthogonal to each other is established. An improved double-power approach law is used to design a terminal sliding mode controller to improve the robot. The integral compensation terms are added to stably compensate the position accuracy of each axis to improve the overall trajectory tracking accuracy, and the cross-coupling control between the axes is used to eliminate the contour error between the axes. It not only weakens the chattering of traditional sliding …

Design And Implementation Of Multi Level Terrain Editing Tool, Xiang Bing, Yang Gang, Xiaohong Dong

Journal of System Simulation

Abstract: 3D terrain is an indispensable scene element in computer games, virtual simulation, geographic information and other applications. Realistic digital terrain plays an important role in enhancing the reality of virtual scene and the immersion of users. Based on the in-depth analysis of terrain features and user modeling requirements, a multi-level terrain editing mode is constructed. In the mode, the terrain editing behavior can be expanded from three levels of point, line and face. Each level has a specific operation target and corresponding editing tools. A multi-level and comprehensive terrain editing tool set composed of point tool, curve tool and …

Simulation And Optimization Of Support Resource Allocation For Multi-Aircraft Large Fleet Re Dispatch, Xu Gang, Zhang Lei, Yonggang Yun, Du Li

Journal of System Simulation

Abstract: In view of the complex combination of support resources of multi-aircraft large fleet re dispatch, based on the new complex factors of different models, different support expectation time and different main resource types, the simulation model of multi-aircraft support process based on queuing network is established and a simulation algorithm for the optimal allocation of support resources is proposed. The algorithm takes the total time of fleet support, the average time of aircraft support, the average intensity of fleet support, the cost of resources, the balance of free and busy support, the balance of resource supply and demand as …

Research On Fingerprint Location Algorithm Based On Dynamic Fingerprint Update, Lizhen Cui, Qiaoli Wang, Qianqian Guo, Yang Yong

Journal of System Simulation

Abstract: According to the characteristics of underground environment, a fingerprint location algorithm based on dynamic fingerprint updating is proposed. FCM(Fuzzy C-Means Clustering) is used to divide the location area according to the signal distribution characteristics, and the training and learning model is established in each sub area. On the basis of FCM algorithm, a HMM(Hidden Markov Model) motion information sequence model based on the location of mobile users is proposed. The dynamic update of fingerprint database is realized by users unconsciously participating in the collection of RSSI(Received Signal Strength Indication) sequence. ANFIS(Adaptive Network-based Fuzzy Inference System) algorithm …

Load System Modeling Of Ultra-Supercritical Coal-Fired Power Unit Based On Improved Particle Swarm Optimization, Yuzhen Sun, Yiwei Tang, Li Shuai

Journal of System Simulation

Abstract: Aiming at the difficulties in modeling due to variables coupling of ultra-supercritical coal-fired power unit and defects in basic particle swarm optimization, an improved particle swarm optimization algorithm for modeling of load system is proposed. The algorithm introduces the idea of adaptive, Cauchy mutation and gradient guidance crossover, which improves the problems of particle swarm optimization algorithm being prone to premature convergence and has the poor local searching ability. By collecting the actual operation data of the power plant, using the adaptive Cauchy mutation and gradient guidance cross particle swarm optimization (GMGPSO) algorithm, the model established and validated …