Digital Commons Network^™

Herd: A Policy-Driven Elastic Resource Distribution Framework For Hpc Deep Learning, Alejandro Guerrero Rodriguez

Open Access Theses & Dissertations

Modern deep learning workloads increasingly rely on distributed computation, and High Performance Computing systems can provide the necessary resources through large GPU allocations across interconnected nodes. Despite this, most distributed training frameworks operate under static resource assignments once a job is deployed. Research on NERSC Perlmutter has shown that 50% of GPU-enabled jobs use 25% or less of available GPU memory, and elastic training can reduce this underutilization by dynamically adjusting active workers. However, existing elastic systems have been developed mainly for cloud environments where fault tolerance and cost optimization are the primary concerns. Applying elastic training to HPC environments …

Ai-Assisted Import Of Static Requirements Documents Into Sysml Models Using A Vibe-Coded Plugin Approach, Joshua Larson

Open Access Theses & Dissertations

Model-based systems engineering (MBSE) promises a single authoritative model in place of static documents, but the translation of requirements documents into that model remains manual, slow, and dependent on scarce modeling expertise. This thesis asks whether an AI-assisted workflow can automate that translation, and whether the tooling itself can be built by vibe coding, in which a practitioner directs an AI in natural language rather than writing code directly. The result is a plugin for Magic Systems of Systems Architect (MSoSA) 2022x that reads a PDF requirements document, extracts requirements through the Anthropic Claude API, and generates a SysML model …

Learning To Unlearn: Unlearning And Meta-Unlearning For Continually Adapting Cybersecurity Threat Detectors, Daniel Lucio

Open Access Theses & Dissertations

Machine learning (ML) models deployed in non-stationary environments must continually adapt to evolving data distributions. This challenge is particularly critical in cybersecurity, where malware, intrusion techniques, and adversarial behaviors evolve over time. Continual learning primarily enables incorporating new knowledge while preserving prior knowledge, however, indiscriminately retaining obsolete and harmful information can hinder future adaptation and consume limited model capacity. We argue that effective adaptation should not only acquire new knowledge, but also selectively discard obsolete and less useful historical knowledge before learning from a new distribution. In this work, we propose a meta-learning framework that learns what to forget to …

Prevention Vs Detection: A Discrete-Event Simulation Assessment Of Inspection Staffing, Pick-To-Light, And Ai-Based Inspection In Aerospace Wire Harness Kitting, Jorge Mares

Open Access Theses & Dissertations

Aerospace wire harness manufacturing operates under conditions that standard industrial engineering methods were not designed to address. Volumes are low, configurations change between orders, assembly is performed manually by certified labor that cannot be expanded quickly, and IPC/WHMA-A-620 Class 3 acceptance requires that every unit be verified rather than sampled. Kit assembly concentrates the resulting risk, because each item retrieved from the rack is an independent selection decision, and the attributes distinguishing a correct item from an incorrect one are frequently not visible to the operator. Errors introduced there are discovered downstream, after substantial certified labor has been invested. A …

Unsupervised Learning For Minimum Error Adaptive Sampling Of Atmospheric Vertical Temperature Profiles, Alejandro Medina

Open Access Theses & Dissertations

Uncrewed aerial systems (UAS) collect atmospheric data on fixed schedules, and endurance limits make blind searching costly. This thesis develops an unsupervised representation that summarizes a multi-decade radiosonde archive into a compact library of atmospheric states, giving a UAS an expectation of the column before it flies. The method standardizes both axes of a profile against the sounding's own surface conditions, which makes the representation independent of season and of station elevation. Applied to 27,270 soundings from Norman, Oklahoma, over the lowest 1.5 km of the atmosphere, 12 representative profiles reconstruct the record to within 1.0 °C of mean absolute …

Electrical Characterization Of Wide-Bandgap Devices Via Laboratory Measurements, Tcad Simulation, And Analytical Modeling, Sebastian Alessandro Mimbela

Open Access Theses & Dissertations

Gallium nitride (GaN) devices are important for high-power, high-frequency, and harsh-environment applications because of their wide bandgap, high breakdown capability, and electrical performance. This thesis focuses on the electrical characterization of GaN-based devices using laboratory measurements, analytical modeling, and TCAD simulation. Current-voltage (I-V) and capacitance-voltage (C-V) measurements were performed on a GaN Schottky diode. The measured data were analyzed to extract key parameters such as ideality factor, Schottky barrier height, series resistance, donor concentration, built-in potential, depletion width, and maximum electric field. A TCAD process was also developed in Synopsys Sentaurus to recreate the main geometry, layer structure, contacts, and …

Hfir Antineutrino Flux Mapping With Scale Mesh Tallies, Jacob Adam Mireles

Open Access Theses & Dissertations

This thesis develops and demonstrates a workflow for mapping the spatial distribution of reactor electron antineutrino flux around the High Flux Isotope Reactor (HFIR) using three-dimensional mesh tallies from the SCALE 6.3.1 Monte Carlo transport suite. A detailed HFIR model is executed with KENO, and the mesh-tally output is post-processed to construct a voxelized antineutrino source term. The source field is normalized to reactor power and propagated to sampled spatial points surrounding the reactor to generate three-dimensional source-side flux maps. The main analysis focuses on how the mesh-based HFIR antineutrino flux differs from an ideal point-source approximation. In the point-source …

Physics Model Calibration Via Functional Kernel Regression, Kwesi Appau Ohene-Obeng

Open Access Theses & Dissertations

Physics based computational simulators encode the governing equations of complex physical phenomena through parameters that must be inferred from empirical observations, and the observational record in modern experimental campaigns is increasingly a high dimensional functional response rather than a scalar summary. The canonical treatment of this calibration problem through the additive discrepancy decomposition of Kennedy and O'Hagan incurs O(N^3obs) Gaussian process inversion costs at functional resolution, sensitivity to discrepancy prior specification, and the confounding identifiability between the calibration parameter and the discrepancy function; scalar reduction strategies bypass the cost but discard the morphological information that the functional response carries about …

A Hybrid Vsm-Des Approach For Bottleneck Identification And Reduction In A Continous One-Piece Flow Assembly System, Ana Sofia Rey Nevarez

Open Access Theses & Dissertations

Identifying bottlenecks in a production line is rarely as easy as pointing to the slowest station and assuming that's where the problem is. The real question is which stations moves the system when you change them. Lean manufacturing has been one of the initiatives that major companies in the U.S. have been trying to adopt to remain competitive. This thesis aims to understand the results of applying a combination of a Discrete-Event Simulation (DES) with a Value Stream Mapping (VSM), to find the bottleneck of a continuous one-piece flow assembly line of a tier-1 power management manufacturer from the El …

From Screening To Risk: Assessment Of Potentially Toxic Elements In Children's Toys Along The El Paso, Tx, U.S.A.-Ciudad Juarez, Chih, Mex Border Region, Jesus Rodriguez-Loya

Open Access Theses & Dissertations

Children are uniquely vulnerable to potentially toxic elements (PTEs), absorbing them more readily than adults and encountering them through the hand-to-mouth behaviors of early childhood. Exposure carries element-specific consequences: lead (Pb) impairs cognitive development with no established safe threshold; cadmium (Cd) damages the kidneys and reduces bone mineralization; arsenic (As) is a recognized carcinogen and developmental neurotoxicant; hexavalent chromium (Cr) is carcinogenic and induces allergic dermatitis; nickel (Ni) is the leading cause of allergic contact dermatitis in children; and excess zinc (Zn) disrupts copper absorption and causes gastrointestinal distress. Toys are a distinctive source of such exposure, yet none had …

A Study On The Complex Doping Of Na2x2teo6 (X = Ni, Cu, Fe, Co, Zn), Omar Salas

Open Access Theses & Dissertations

Na2Ni2TeO6 is a layered honeycomb antiferromagnet in which Ni2+ ions form magnetic honeycomb planesseparated by sodium layers. In this work, four compositionally complex oxides derived from Na2Ni2TeO6 were synthesized to test whether the NNTO-type structure remains stable when 50% of the Ni sublattice is replaced by mixtures of Cu, Zn, Fe, and Co. Powder X-ray diffraction and neutron powder diffraction show that all four samples retain the hexagonal P 63/mcm average structure of the parent NNTO phase, demonstrating that the layered honeycomb framework is stable against substantial multiple-cation substitution. A small CoFe2O4 secondary phase is present in all four samples …

Towards A Multi-Framework Approach To Explainable Artificial Intelligence, Henry Salgado

Open Access Theses & Dissertations

Artificial Intelligence (AI) and machine learning (ML) models are increasingly being deployed to support decision-making in high-stakes domains such as healthcare, criminal justice, and education, where trust, accountability, and transparency are critical. However, increasing model complexity has made many modern systems insufficiently transparent. Existing approaches to explainable AI (XAI) typically emphasize either intrinsic model simplicity or post-hoc attribution methods that estimate feature importance for predictions. While these approaches provide valuable insights into model behavior, they do not necessarily establish whether the identified importance is grounded in the underlying data patterns or in the structural relationships that generate model behavior. Many …

Integrated Framework For Tsn-Enabled Ot Networks And Scalable Edge Computing To Enable Real-Time Feedback Loop, Taposh Kumer Sarker

Open Access Theses & Dissertations

The advent of Industry 5.0 envisions smart manufacturing characterized by human centricity, sustainability, and systemic resilience. Realizing this vision requires the seamless convergence of Information Technology (IT) and Operational Technology (OT) networks. However, integrating massive, stochastic IT edge computing workloads with deterministic physical control loops introduces severe architectural friction, inherently threatening the safety guarantees required by industrial machinery. To resolve this fundamental incompatibility, this dissertation proposes the Edge-Augmented Real-Time Industrial Control System (EA-RICS).

EA-RICS is a comprehensive, multi-layered architecture designed to dismantle systemic bottlenecks across the physical data plane, the centralized control plane, and the edge operating system. First, the …

2026 August News Report, Morehead State University. Office Of Communications & Marketing.

Communications and Marketing Publications Archive

The August 2026 Morehead State University News Report.

Summer 2026 - Full Issue, Georgia Library Quarterly

Georgia Library Quarterly

No abstract provided.

Development And Evaluation Of Alternative Mandrel Materials For The Fabrication Of In-House Silicone Vascular Models, Cole Capitani

Master's Theses

Dr. Kristen Cardinal’s Tissue Engineering Laboratory at California Polytechnic State University, San Luis Obispo develops tissue-engineered blood vessel mimics (BVMs) for preclinical testing of neurovascular devices. These BVMs are reliant on silicone vascular models that define the vessel geometry, support endothelial cell attachment, and allow devices to be evaluated in a controlled in vitro environment. Previous work in the lab established injection molding as a viable method for fabricating silicone vascular models and utilized acrylonitrile butadiene styrene (ABS) for sacrificial mandrels. However, this workflow relied on the use of acetone for mandrel dissolution, required single-use mandrels, had extended post-processing times, …

When Does Emigration Reduce Domestic Economic Activities? Evidence On The Loss Of Management Capacity From A Visa Lottery, Marup Hossain, Ahmed Mushfiq Mobarak

Discussion Papers

Economic development is often conceived as structural transformation and migration out of rural agriculture (Lewis, 1954; Fei & Ranis, 1964). Understanding development therefore requires us to identify how rural economies transform in response to emigration. We study how international migration reshapes domestic economic activity in origin households using a randomized visa lottery that gave Bangladeshi men job opportunities in Malaysia. Winning the lottery raises total household income through remittances, but income earned in Bangladesh declines due to contractions in nonfarm business activity. In contrast, crop income remains stable despite large reductions in agricultural input expenditures, because farming gets delegated through …

Secure And Compassionate Dementia Care: Non-Intrusive Remote Monitoring Of Falls, Wandering, And Agitation, Awan-Ur- Rahman

Master's Theses

Alzheimer’s disease and related dementias (ADRD) present significant safety challenges, as affected individuals may experience falls, wandering, agitation, and a progressive decline in independence. Although continuous monitoring can enable timely intervention, many existing systems rely on cameras or wearable devices, which may introduce concerns related to privacy, comfort, and sustained use. This thesis explores privacy-preserving and non-intrusive remote monitoring through two complementary studies. The first study presents an ambient Wi-Fi channel state information framework for recognizing agitation and eight daily activities. The proposed approach translates behavioral and physiological markers commonly captured by wearable sensors into the Wi-Fi sensing domain. The …

Constructing Female High School Students’ Stem Self-Efficacy: The Role Of High School Stem Research Experiences, Hyunjin Son

Dissertations

In the U.S., nearly a quarter of the workforce is employed in Science, Technology, Engineering, and Mathematics (STEM) occupations. However, the STEM workforce lacks diversity, limiting perspectives to address real-world problems facing society. While many factors influence decisions to pursue STEM majors in college and enter the STEM workforce, one factor commonly associated with performance and persistence in STEM is STEM self-efficacy. Independent research experiences have been shown to increase conceptual understanding and confidence in STEM, leading to STEM degrees and careers. As high school co-curricular STEM research opportunities become more common, the effects of such experiences on students’ STEM …

Secondary Biology Teachers’ Understandings Of Growth Mindset, Kathryn Chapman

Dissertations

This case study investigates how high school biology teachers understand and apply growth mindset principles in their daily instruction. Using surveys, semi‑structured interviews, and focus groups, the study examines how biology teachers interpret mindset, how they attempt to enact it, and how curriculum demands shape their efforts. Findings show that teachers’ beliefs about intelligence strongly influence their feedback, expectations, and instructional decisions. Teachers described biology as a subject in which struggle is highly visible, emotionally charged, and often tied to students’ prior knowledge, making mindset work both necessary and challenging. Results also indicate that students’ mindsets are shaped by broader …

Machine Learning For Predictive Energy And Emissions Modeling Of Vehicles And Power Grids In The United States, S M Tanvir Faysal Alam Chowdhoury

Dissertations

The environmental benefits of electric vehicle (EV) adoption depend on more than replacing internal combustion engine vehicles with electric powertrains. EV adoption reshapes electricity demand, interacts with regional generation mixes, and influences travel behavior and congestion, creating a coupled transportation-energy system in which vehicle and power-plant emissions must be evaluated together. This dissertation develops machine-learning frameworks for predicting energy consumption and emissions from vehicles and power grids under rising EV adoption. The first component forecasts grid emissions from EV charging. Using simulation data from NREL's Cambium database, a Prophet-based time-series framework predicts carbon dioxide, nitrous oxide, and methane emission rates …

Purdue University General Institutional Description, Purdue University Office Of Research

University General Facility Boilerplate Descriptions

Overview of Purdue University mission, statistics, and key cross-campus infrastructure.

Protecting Wetlands In Georgetown County, Sophia Harrison

Goal 6: Clean Water and Sanitation

No abstract provided.

Pseudo-Riesz Bases, Deborpita Biswas

All Dissertations

In this dissertation, we will introduce the concept of pseudo-Riesz bases, extending the influential notion of near-Riesz bases first proposed by J. Holub in the 1990s. Our goal is to develop an analogous theory for pseudo-Riesz bases, obtaining results parallel to those known for near-Riesz bases, with suitable modifications.

We will develop the foundational theory of pseudo-Riesz bases. Unlike classical Riesz bases, these sequences need not be complete or independent; however, they still retain meaningful expansion properties. We will investigate the conditions under which a Bessel sequence can be transformed into a Riesz basis through finite modifications. In particular, we …

Exploring Public Perceptions Of Genetic Risk In Substance Use Disorders, Brooke A. Goodenow

All Dissertations

This dissertation examines genetic risk perception, with specific emphasis on substance use disorders (SUDs), through a collection of three manuscripts. The dissertation examines genetic risk perception for hereditary diseases in existing literature, describes a methodological framework for capturing genetic risk perception across social media discourse, and identifies themes and emotional tone associated with genetic risk perception for SUDs across two major social media platforms (X and Reddit).

Combined, these three manuscripts expand and support previous findings about genetic risk perception and its role in shaping health beliefs and associated behaviors. The dissertation findings reveal numerous similarities in genetic risk perception …

Broadening Participation In Computing: Retention Pathways For Supporting Underrepresented Students, Oluwafemi Osho

All Dissertations

The computing sector's rapid growth has intensified demand for computing professionals, yet undergraduate computer science (CS) programs continue to lose substantial numbers of students, often within the first two years, before they reach degree completion. Attrition is not evenly distributed: women and other underrepresented minority (URM) students remain less likely to persist in CS despite broader increases in enrollment. These persistent disparities suggest the need to understand not only how to broaden entry into computing but also how students' experiences within programs shape retention over time. In this dissertation, I address this critical challenge by focusing on identifying inclusive strategies …

Three-Dimensional Gaussian Reconstruction Of Large-Scale Scenes Under Multi-View Geometry Constraints, Haohao Cui, Yanqiang Di, Qing Liu, Xianguo Meng

Journal of System Simulation

Abstract: To enhance the geometry reconstruction quality of the GS algorithm in large-scale scene reconstruction, an optimization method constrained by multi-view geometry reconstruction results was proposed. 2D Gaussian planes were used as geometric primitives to overcome depth anisotropy, and dense depth maps generated by DUSt3R and aligned by sparse point clouds were introduced as constraints. By designing a multi-stage optimization strategy that decouples geometry and rendering, the gradient conflict problem in multi-objective training was solved. Experiments on the MatrixCity dataset indicate that the method surpasses comparison methods in related indicators of geometry reconstruction quality and rendering quality in large-scale scenes. …

Interplay Of Atomicity, Polarization, And Quantization On Performance And Reliability In Gan Finfets, Salim I. Almenshad

Dissertations

GaN based nanoscale high-electron-mobility transistors (HEMTs) are the next generation of transistor technology that features the unique combination of higher power, wider bandwidth, low noise, higher efficiency, and temperature/radiation hardness than conventional AlGaAs and Si based technologies. However, enhancement mode operation for nanoscale channel region is the most recent needed area of research. As well as reliability of these devices remains a major concern. The main objective of this proposed research is to numerically investigate the role of atomistic inhomogeneity and quantization in low-dimensionality lateral tri-gated AlGaN/GaN FinFET devices. It is shown that 3-D configured AlGaN/GaN FinFETs with narrow enough …

Applications Of Machine Learning For Nondestructive Evaluation Data Analysis, Connor Chase Seavers

Dissertations

A fourth industrial revolution is currently taking place in the manufacturing sector, called “Industry 4.0,” where material production and fabrication processes are becoming fully automated using integrated sensors, closed-loop feedback mechanisms, and artificial intelligence/machine learning (AI/ML). The nondestructive evaluation (NDE) field has begun to follow suit by integrating AI/ML techniques in material inspection processes to automate defect detection for many NDE methods. Since material inspection processes are paramount in ensuring safety and reliability in critical manufactured products, it is important that new AI/ML techniques are trustworthy for NDE data analysis. The current approach to building trust in new AI/ML methods …

Global Kinetic Dynamics Of Black Hole Magnetospheres And Accretion, Martin Luepker

Arts & Sciences Graduate Student Theses and Dissertations

The Event Horizon Telescope (EHT) has provided stunning images of the immediate vicinity of two supermassive black holes, M87* and Sgr A*, revealing the plasma swirling around them. To understand the complex emission observed, researchers have turned to general relativistic magnetohydrodynamic (GRMHD) simulations, which model the plasma as a fluid. These simulations have been successful in reproducing some features of the EHT images, but they rely on assumptions that do not apply in the low-density, high-magnetic-field environments around these black holes. General relativistic particle-in-cell (GRPIC) simulations offer a first-principles approach to modeling this environment, but their application to global black …