Physical Sciences and Mathematics Commons^™

Emergent Capabilities Of Llms For Software Engineering, Conor O'Brien

Dissertations, Theses, and Masters Projects

A growing interest for Large Language Models (LLMs) is how increasing their size might result in changes to their behavior not predictable from relatively smaller-scaled models. Analyzing these emergent capabilities is therefore crucial to understanding and developing LLMs. Yet, whether LLMs exhibit emergence, or possess emergent capabilities, is a contested question. Furthermore, most research into LLM emergence has focused on natural language processing tasks and models suited for them. We focus on investigating emergence in the context of software engineering, and recontextualize the discussion of emergence in the context of prior research. We propose a multifaceted pipeline for evaluating and …

Automated Bug Report Management To Enhance Software Development, Yang Song

Dissertations, Theses, and Masters Projects

Bug report management is crucial yet challenging process that affects the efficiency of software development process. It involves reporting, triaging, detecting duplicates, assigning, localizing, fixing bugs, and thorough verification. The high volume and variety of bug reports complicate these tasks, highlighting the need for innovative solutions to improve the process and boost development efficiency. This dissertation explores the potential of automating the bug management process to optimize the effectiveness of software development and maintenance. It focuses on three key stages of bug management: reporting, assignment, and localization, presenting four innovative solutions for these phases. First, it discusses the challenges faced …

Investigating The Latent Security And Privacy Risks In Consumer-Oriented Software Systems, Kaushal Kafle

Dissertations, Theses, and Masters Projects

Consumer-oriented software systems have become the foundation on which consumer data is collected and transported from the consumers to the data processors. They are complex, with various interconnected, heterogeneous components working together, making their security and privacy analysis challenging, and impact on the user uncertain. In this work, we first explore how security threats can arise in novel context in such systems by performing a security evaluation of data-store based smart home platforms and the overall security risks posed by the design of routines within such platforms. We analyze various components of the smart home such as the platform’s permission …

Transitions To Two-Hadron States From Quantum Chromodynamics, Felipe Ortega Gama

Dissertations, Theses, and Masters Projects

The rich spectrum of hadrons reflects the complexity of interactions between quarks and gluons confined within them. Most of these hadrons are extremely short-lived and are called resonances. Experimentally, they are observed indirectly through their effects on the energy distribution in scattering experiments. Additionally, the non-perturbative nature of Quantum Chromodynamics (QCD), which governs the dynamics of quarks and gluons, prevents the implementation of known analytical techniques for calculating transition and interaction rates between hadrons. Lattice QCD (LQCD), a numerical implementation of QCD, provides a non-perturbative approach to studying the spectrum, as long as we understand how to account for finite-volume …

Probing With Displacements For Variance Reduction And The Effectiveness Of Sketched Krylov Eigenvalue Solvers, Heather Maria Switzer

Dissertations, Theses, and Masters Projects

Scientific Computing is a multidisciplinary field that intersects Computer Science, Mathematics, and some other discipline to address complex problems utilizing computational systems. Numerical Linear Algebra (NLA), a subfield within Scientific Computing, develops and analyzes numerical algorithms for tasks involving linear operators such as matrices and their transformations. This dissertation focuses on developing kernels for two specific areas of NLA. Firstly, we explore variance reduction methods for trace approximation. In Lattice Quantum Chromodynamics (LQCD), computing the trace of a matrix inverse is crucial for investigating interactions among quarks and gluons in subatomic space. However, directly computing a matrix inverse is computationally …

Aquaculture Governance: Five Engagement Arenas For Sustainability Transformation, Darien D. Mizuta, Stefan Partelow

VIMS Articles

A greater focus on governance is needed to facilitate effective and substantive progress toward sustainability transformations in the aquaculture sector. Concerted governance efforts can help move the sector beyond fragmented technical questions associated with intensification and expansion, social and environmental impacts, and toward system-based approaches that address interconnected sustainability issues. Through a review and expert-elicitation process, we identify five engagement arenas to advance a governance agenda for aquaculture sustainability transformation: (1) setting sustainability transformation goals, (2) cross-sectoral linkages, (3) land–water–sea connectivity, (4) knowledge and innovation, and (5) value chains. We then outline the roles different actors and modes of …

Northern Neck Regional Shallow Draft Channel Dredging Plan: Feasibility Phase, Donna A. Milligan, Cameron W. Green, C. Scott Hardaway Jr., Christine A. Wilcox, Claire M. Rae, Scott Lerberg, Alex Demeo, George Brooks, Mark Mansfield

Reports

The Initial Phase of the project (Phase 1), used remote sensing and other data collection to develop a method for determining which waterbodies on the Northern Neck of Virginia need dredging. The analysis included most of the waterbodies in the counties of Lancaster, Northumberland, Richmond, and Westmoreland. From this analysis, 19 waterbodies emerged as potentially needing dredged based on physical parameters and residential and economic usage (Milligan et al., 2023).

In this Feasibility Phase (Phase 2), more detailed site data were collected to provide data to the localities for consideration. These tasks were included in the analysis:

1. Historic shore …

Northern Neck Regional Shallow Draft Channel Dredging Plan: Initial Phase, Donna A. Milligan, Grace M. Massey, C. Scott Hardaway Jr., Christine A. Wilcox, Cameron W. Green, Mark Mansfield

Reports

The purpose of this project is to develop a regional dredging program for the localities of the Northern Neck. This report encompasses Phase 1 of the project, which included creating a database of waterbodies in Lancaster, Northumberland, Richmond, and Westmoreland Counties and the Town of Colonial Beach along the Potomac and Rappahannock Rivers and Chesapeake Bay. Generally, these waterbodies can be categorized into three basic types of shallow draft channels: federally-authorized to include aids to navigation (ATONS), non-federal with ATONS, and non-federal without ATONs. Along the Northern Neck, presently identified, are 13 federal channels, 37 non-federal channels in creeks with …

The Development Of Strategies For Coastal Wetland Conservation Prioritization In Virginia Under Climate Change, Molly Mitchell, Z. Lv, Bryan Watts, Chance Hines, Kory Angstadt, David Stanhope, Kirk J. Havens, Donna Marie Bilkovic, Karinna Nunez

Reports

This project focuses on developing strategies to improve our understanding and strengthen the sustainability of Virginia’s coastal wetlands to climate change impacts. Marsh migration under sea level rise is a primary pathway for marsh persistence. However, the resulting extent and habitat function of the newly migrated marsh is still being explored. In addition, accelerating sea level rise suggests that coastal wetlands need frequent monitoring to assess change. This project addresses Virginia regional priority 1, to develop a framework to overcome barriers to use existing wetland assessment methodologies and studies for restoration/compensatory mitigation projects to improve functional performance of aquatic resources …

Parameter Estimation For Patient Enrollment In Clinical Trials, Junyan Liu

Undergraduate Honors Theses

In this paper, we study the Poisson-gamma model for recruitment time in clinical trials. We proved several properties of this model that match our intuitions from a reliability perspective, did simulations on this model, and used different optimization methods to estimate the parameters. Although the behaviors of the optimization methods were unfavorable and unstable, we identified certain conditions and provided potential explanations for this phenomenon and further insights into the Poisson-gamma model.

Crystal Structure And Hirshfeld Surface Analysis Of Cyclo-Tetrabromido-1k2br,3k2br-Tetrakis(Pi2-2-{[(Pyridin-2-Yl)Methyl]Amino}Ethane-1-Thiolato-Krystal Structure And Hirshfeld Surface Analysis Ofcyclo-Tetrabromido-1j2br,3j2br-Tetrakis(L2-2-{[(Pyridin-2-Yl)Methyl]Amino}Ethane-1-Thiolato-K3n,S:S)Tetramercury(Ii), I. D. Thomas, K. R. Kocher, J. A. Viehweg, Robert Pike, Deborah Bebout

Arts & Sciences Articles

The macrometallacyclic title compound, [Hg4Br4(C8H11N2S)4] or [((HgL2)(HgBr2))2] (1) where HL = 2-{[(pyridin-2-yl)meth­yl]amino}­ethane-1-thiol, was prepared and structurally characterized. The Hg2+ complex crystallizes in the P21/c space group. The centrosymmetric Hg4S4 metallacycle is constructed from metal ions with alternating distorted tetra­hedral Br2S2 and distorted seesaw N2S2 primary coordination environments with pendant pyridyl groups. The backfolded extended chair metallacycle conformation suggests inter­actions between each of the bis-chelated mercury atoms and Br atoms lying above and below the central Hg2S4 plane. Supra­molecular inter­actions in 1 include a fourfold aryl embrace and potential hydrogen bonds with bromine as the acceptor. Hirshfeld surface analysis indicates that …

Results For The 2023 Vims Industry Cooperative Surveys Of The Mid-Atlantic Bight And Georges Bank Scallop Resource Areas, Sally Roman, David Rudders, Carter Lin, Edwin D. Sánchez

Reports

The Virginia Institute of Marine Science (VIMS) conducted high resolution sea scallop dredge surveys of the Mid-Atlantic Bight (MAB) and Georges Bank (GB) scallop resource areas during May–July 2023.

Dataset: Marsh Migration Methodology Development For Wetland Restoration Targeting, Molly Mitchell, Karinna Nunez, Christine Tombleson, Julie Herman

Data

Coastal marsh loss is a significant issue globally, due in part to rising sea levels and high levels of coastal human activity. Marshes have natural mechanisms to allow them to adapt to rising sea levels, however, migration across the landscape is one of those mechanisms and is frequently in conflict with human use of the shoreline. Ensuring the persistence of marshes into the future requires an understanding of where marshes are likely to migrate under sea level rise and targeting those areas for conservation and preservation activities. The goal of this project was to 1) compile existing datasets and information …

Synthesis Of Shoreline, Sea Level Rise, And Marsh Migration Data For Wetland Restoration Targeting Final Report, Molly Mitchell, Karinna Nunez, Christine Tombleson, Julie Herman

Data

Coastal marsh loss is a significant issue globally, due in part to rising sea levels and high levels of coastal human activity. Marshes have natural mechanisms to allow them to adapt to rising sea levels, however, migration across the landscape is one of those mechanisms and is frequently in conflict with human use of the shoreline. Ensuring the persistence of marshes into the future requires an understanding of where marshes are likely to migrate under sea level rise and targeting those areas for conservation and preservation activities. The goal of this project was to 1) compile existing datasets and information …

Assessment Of Changes Of Complex Shoreline From Medium‑Resolution Satellite Imagery, Nikolay P. Nezlin, Julie Herman, Jonathan Hodge, Stephen Sagar, Robbi Bishop-Taylor, Guangming Zheng, John M. Digiacomo

VIMS Articles

The imagery collected by medium-resolution earth-observing satellites is a powerful and cost-effective tool for the quantitative assessment of shoreline dynamics for water bodies of different spatial scales. In this study, we utilize imagery collected in 1984–2021 on the Middle Peninsula, Virginia, bordering the Chesapeake Bay, USA, by medium-resolution (10–30 m) satellites Landsat-5/7/8 and Sentinel-2A/B. The data was managed in the Earth Analytics Interoperability Lab (EAIL) Data Cube built and configured by the Commonwealth Scientific and Industrial Research Organization (CSIRO, Australia and Chile). The assessments of shoreline change demonstrate adequate agreement with assessments based on aerial photography collected during 1937–2009 by …

A Language Framework For Modeling Social Media Account Behavior, Alexander C. Nwala, Alessandro Flammini, Filippo Menczer

Arts & Sciences Articles

Malicious actors exploit social media to inflate stock prices, sway elections, spread misinformation, and sow discord. To these ends, they employ tactics that include the use of inauthentic accounts and campaigns. Methods to detect these abuses currently rely on features specifically designed to target suspicious behaviors. However, the effectiveness of these methods decays as malicious behaviors evolve. To address this challenge, we propose a language framework for modeling social media account behaviors. Words in this framework, called BLOC, consist of symbols drawn from distinct alphabets representing user actions and content. Languages from the framework are highly flexible and can be …

Identification Of Transient Radio Frequency Occlusion Events In Urban Environments, Margaret M. Rooney, Mark Hinders

Arts & Sciences Articles

We model the propagation of SHF OFDM signals around vehicles and buildings since these are the most common elements present in urban environments that could lead to complex radio frequency signal scattering. Scenarios involving temporary hidden node situations, which we term transient occlusion events, are simulated and compared to scenarios where a line of sight transmission event occurs. Sets of fingerprints generated from signals recorded in full-wave 3D finite difference time domain simulations of these two different types of situations are compared, and features in the fingerprints corresponding to the occlusion of a transmitted signal by a vehicle or a …

Gis Data: Caroline County, Maryland – Shoreline Inventory Data 2023, Karinna Nunez, Tamia Rudnicky, Sharon Killeen, Jessica Hendricks, Catherine R. Duning, Evan Hill

Data

The shoreline inventory files have been generated to support the application of the Maryland Shoreline Stabilization Model (SSM), developed by the Center for Coastal Resources Management (CCRM), Virginia Institute of Marine Science (VIMS), to enhance and streamline regulatory decision making in Maryland. This shoreline inventory includes the features needed as inputs to run the SSM.

The data developed for the Shoreline Inventory is based on a three-tiered shoreline assessment approach. This assessment characterizes conditions by using observations made remotely at the desktop using high resolution imagery. The three-tiered shoreline assessment approach divides the shore zone into three regions:

1) the …

Gis Data: Prince George’S County, Maryland – Shoreline Inventory Data 2023, Karinna Nunez, Tamia Rudnicky, Sharon Killeen, Jessica Hendricks, Catherine R. Duning, Evan Hill

Data

The shoreline inventory files have been generated to support the application of the Maryland Shoreline Stabilization Model (SSM), developed by the Center for Coastal Resources Management (CCRM), Virginia Institute of Marine Science (VIMS), to enhance and streamline regulatory decision making in Maryland. This shoreline inventory includes the features needed as inputs to run the SSM.

The data developed for the Shoreline Inventory is based on a three-tiered shoreline assessment approach. This assessment characterizes conditions by using observations made remotely at the desktop using high resolution imagery. The three-tiered shoreline assessment approach divides the shore zone into three regions:

1) the …

Gis Data: Harford County, Maryland – Shoreline Inventory Data 2023, Karinna Nunez, Tamia Rudnicky, Sharon Killeen, Jessica Hendricks, Catherine R. Duning, Evan Hill

Data

The shoreline inventory files have been generated to support the application of the Maryland Shoreline Stabilization Model (SSM), developed by the Center for Coastal Resources Management (CCRM), Virginia Institute of Marine Science (VIMS), to enhance and streamline regulatory decision making in Maryland. This shoreline inventory includes the features needed as inputs to run the SSM.

The data developed for the Shoreline Inventory is based on a three-tiered shoreline assessment approach. This assessment characterizes conditions by using observations made remotely at the desktop using high resolution imagery. The three-tiered shoreline assessment approach divides the shore zone into three regions:

1) the …

Gis Data: Cecil County, Maryland – Shoreline Inventory Data 2023, Karinna Nunez, Tamia Rudnicky, Sharon Killeen, Jessica Hendricks, Catherine R. Duning, Evan Hill

Data

The shoreline inventory files have been generated to support the application of the Maryland Shoreline Stabilization Model (SSM), developed by the Center for Coastal Resources Management (CCRM), Virginia Institute of Marine Science (VIMS), to enhance and streamline regulatory decision making in Maryland. This shoreline inventory includes the features needed as inputs to run the SSM.

The data developed for the Shoreline Inventory is based on a three-tiered shoreline assessment approach. This assessment characterizes conditions by using observations made remotely at the desktop using high resolution imagery. The three-tiered shoreline assessment approach divides the shore zone into three regions:

1) the …

A Satellite Imagery Approach To Estimating Migratory Flows In Guatemala Using Convolutional Neural Networks, Sarah Larimer

Undergraduate Honors Theses

Being able to predict migratory flows is important in ensuring political, social, and economic stability. In the wake of violence, unrest, natural disasters, and social pressures, millions of mi- grants have fled Central America in search of a better life. However, due to the infrequent nature and high cost of census data, there is a need for a more remote and up to date approaches. Con- volutional Neural Networks offer a computer vision based approach that is cheaper and with significantly less lag. In this study, we seek to evaluate the effectiveness of different convolu- tional neural networks in predicting …

Pion Detection For The Moller Parity-Violating Electron Scattering Experiment, Michael Tristan Hurst

Undergraduate Honors Theses

The MOLLER Experiment at Jefferson Lab intends to make a precise measurement of the weak charge of the electron through parity-violating electron scattering. To achieve the level of precision required for the experiment, background rates of events other than electron-electron scattering must be known. Working with data from Monte-Carlo simulations created using a GEANT4 simulation package, I show that the combined signal from two existing detector subsystems of the MOLLER experiment allow for particle identification between electron and pion events. I worked to optimize an additional ‘Pion Exit Scintillator’ which improves the ability to distinguish particle identity at the cost …

Identifying Social Media Users That Are Susceptible To Phishing Attacks, Zoe Metzger

Undergraduate Honors Theses

Phishing scams are a billion-dollar problem. According to Threatpost, in 2020, business email compromise phishing attacks cost the US economy $ 1.8 billion. Social media phishing scams are also on the rise with 74% of companies experiencing social media attacks in 2021 according to Proofpoint. Educating users about phishing scams is an effective strategy for reducing phishing attacks. Despite efforts to combat phishing, the number of attacks continues to rise, likely indicative of a reticence of users to change online behaviors. Existing research into predicting vulnerable social media users that are susceptible to phishing mostly focuses on content analysis of …

Fecal Pellet Production By North Atlantic Zooplankton, Michael Gibson

Undergraduate Honors Theses

Fecal pellet carbon (FPC) production by zooplankton is a significant component of the ocean’s biological carbon pump: the suite of biological processes that mediate export of carbon to the deep ocean, ultimately leading to the sequestration of atmospheric carbon dioxide in the ocean. In this study, mesozooplankton (zooplankton 0.2 mm to ~2 cm) were collected from the epipelagic zone in the temperate North Atlantic Ocean during day and night in May 2021. Zooplankton were live separated into five size fractions and incubated on board ship in natural surface seawater to measure fecal pellet production rate of the mixed mesozooplankton community. …

A Study Of Reciprocal Underwater Motion And Its Use In Algae Harvesting, Marguerite Bright

Undergraduate Honors Theses

In 2009, many research groups at different companies and universities were funded by Statoil to study the use of algae as a potential biofuel. Combined with the Chesapeake Bay TMDL given by the EPA, a team at William & Mary and VIMS studied the growth and harvest of wild algae in the York River. This method also removed harmful nutrients such as nitrogen and phosphorus from the waterways. Other independent research projects stemmed from this. In 2014, a research team sought to commercialize and automate the IWAGS system, and found that a single oscillating blade was the most effective. This …

Power Profiling Smart Home Devices, Kailai Cui

Undergraduate Honors Theses

In recent years, the growing market for smart home devices has raised concerns about user privacy and security. Previous works have utilized power auditing measures to infer activity of IoT devices to mitigate security and privacy threats.

In this thesis, we explore the potential of extracting information from the power consumption traces of smart home devices. We present a framework that collects smart home devices’ power traces with current sensors and preprocesses them for effective inference. We collect an extensive dataset of duration > 2h from 6 devices including smart speakers, smart camera and smart display. We perform different classification tasks …

Kfactorvae: Self-Supervised Regularization For Better A.I. Disentanglement, Joseph S. Lee

Undergraduate Honors Theses

Obtaining disentangled representations is a goal sought after to make A.I. models more interpretable. Studies have proven the impossibility of obtaining these kinds of representations with just unsupervised learning, or in other words, without strong inductive biases. One strong inductive bias is a regularization term that encourages the invariance of factors of variations across an image and a carefully selected augmentation. In this thesis, we build upon the existing Variational Autoencoder (VAE)-based disentanglement literature by utilizing the aforementioned inductive bias. We evaluate our method on the dSprites dataset, a well-known benchmark, and demonstrate its ability to achieve comparable or higher …

Black Hole Entropy In Ads/Cft And The Schwinger-Keldysh Formalism, Luke Mrini

Undergraduate Honors Theses

The Schwinger-Keldysh formalism for non-equilibrium field theory provides valuable tools for studying the black hole information loss paradox. In particular, there exists a Noether-like procedure to obtain the entropy density of a system by a discrete Kubo-Martin-Schwinger (KMS) variation of the action. Here, this Noether-like procedure is applied to the boundary action of an asymptotically anti-de Sitter (aAdS) black hole spacetime in maximally extended Kruskal coordinates. The result is the Kubo formula for shear viscosity, which is known in theories with an Einstein gravity dual to have a universal, constant ratio with the entropy density and is proportional to the …

Materials Characterization For Microwave Atom Chip Development, Jordan Shields

Undergraduate Honors Theses

This thesis describes research to characterize materials to be implemented on a microwave atom trap chip, which will be able to trap and spatially manipulate atoms using the spin-specific microwave AC Zeeman effect. Potential applications of this research include atom-based interferometry and quantum computing.

Namely, this thesis describes the characterization of the following: (1) the dielectric constant of a well-characterized substrate, Rogers RO4350B, in order to provide proof-of-concept for a method that can be applied to the chip’s substrate, aluminum nitride (AlN), (2) the maximum current that will be able to be applied to the chip, and (3) surface roughness …