Physical Sciences and Mathematics Commons^™

Exploring Loss Functions In Machine Learning, Yujie Wang

CGU Theses & Dissertations

The loss function plays a critical role in machine learning. It is fundamental in training, evaluating, and optimizing machine learning models, directly impacting their effectiveness and efficiency in solving specific tasks. We explore three new loss functions and their applications. Softmax Cross-Entropy Loss, stands as a prevalent choice in neural network classification tasks. It treats all misclassifications uniformly. However, multi-class classification problems often have many semantically similar classes. We should expect that these semantically similar classes will have similar parameter vectors. We introduce a weighted loss function, the tree loss as a drop-in replacement for the cross entropy loss. The …

Post Polymerization Modification And Preparation Of Dynamic Polymer Networks Via Guanylation Of Aryl Carbodiimides, Conner J. Klingler

WWU Graduate School Collection

Styrenic carbodiimide (CDI) polymers have shown an interesting ability to form reversible polymer networks, known as covalent adaptable networks (CANs), through an uncatalyzed reaction with multifunctional amines. CANs have been a topic of interest in polymer chemistry due to their ability to be reprocessed, which is not found in classic thermoset polymer networks. However, not much is known about the capabilities of the nucleophilic addition of amines to aryl CDI repeat unit structures that goes into making these networks due to the understudied reactions involved. This thesis aims to develop a better understanding of the chemistry involved through various post …

Reconstructing Vegetation Cover During The Retreat Of The Puget Lobe Of The Cordilleran Ice Sheet Using Organic Matter From Marine Sediment Cores, Emily Mcquarrie

WWU Graduate School Collection

The dynamics of the Cordilleran Ice Sheet (CIS) over the last glacial period are complex, with several instances of ice retreat and readvance before the final retreat. This study demonstrates that in western Washington, marine sediment cores can be used to reconstruct continental ice cover. Here, I present a reconstruction of terrestrial landscape evolution within the Puget Sound during the last deglaciation, using geochemical analyses of organic matter from marine sediment cores. I measured the carbon isotope composition in bulk organic matter (d¹³C_org) and concentrations of leaf wax biomarkers (n-alkanes and fatty acid methyl …

Low-Rank Matrix And Tensor Models For Data Science Applications, Jeremy Moulton Myers

Dissertations, Theses, and Masters Projects

Low-rank approximations play an important role in data science analysis and applications. When the model is linear and the data is represented as a matrix, the optimal rank-r approximation is given by the r dominant factors of the singular value decomposition (SVD). When the model is multilinear and the data is represented as a multi-way array called a tensor, the optimal rank-r approximation may not even exist. Nonetheless, the canonical polyadic decomposition (CP) provides a useful low-rank tensor approximation for interpretability and analysis in a way similar to the matrix SVD but across multiple axes simultaneously. In this dissertation, we …

Environmental Controls On Carbon And Nitrogen Cycling In Alaskan Arctic Coastal Lagoons, Brian Seung Taek Kim

Dissertations, Theses, and Masters Projects

At the intersection of land, sea, and atmosphere, coastal lagoons act as bioreactors, processing both terrestrial and autochthonous organic matter production before export to the coastal ocean. Approximately half of the Beaufort Sea coast is outlined by barrier island chains that enclose shallow lagoons and sounds. These lagoons are subject to extreme seasonal variations in ice cover, temperature, and salinity, yet are home to a diverse and productive food web. Bound between the arctic tundra and Beaufort Sea, these shallow systems receive and process resources from both sea and particularly land in the form of coastal erosion and riverine export. …

Investigating Uncertainty In Gaussian Process Models, Wilson Strasilla

Master's Projects

N. A.

Revegetation Of Two Drained Reservoirs On The Elwha River Ten Years After Dam Removal, Michael Trier

EWU Masters Thesis Collection

The Elwha and Glines Canyon dams on the Elwha River, Washington, USA were removed between 2011 and 2014 after being in place for a century, and at the time were the largest dams ever removed. The resulting dewatering of Lake Aldwell, the reservoir of the Elwha dam, and Lake Mills, the reservoir of the Glines Canyon dam exposed over 270 hectares of land that had been submerged. While several studies have examined the revegetation of these drained reservoirs, none have gone beyond 5 years since dam removal. These studies found that sediment grain size and reservoir landform (valley walls, terraces …

Dam Dimensions And Surface Porosity Affect The Water Storage Capacity Of Beaver Dam Analogs Compared To Natural Beaver Dams, Sawyer Nagle

EWU Masters Thesis Collection

Widespread stream incision in the Western United States, exacerbated by climate change and anthropogenic activities, necessitates effective restoration strategies. My study compares water retention by Beaver Dam Analogs (BDAs) to natural beaver dams and undammed control reaches. I compared how dam dimensions affect the hydraulic performance of both BDAs and beaver dams, revealing gaps in research on their comparative effectiveness. My study was conducted across eight watersheds in Washington state and two in Idaho. Hydraulic retention time (HRT), water travel time, and pool volume were response variables used to determine dam effectiveness. Dam height, thickness, length and porosity were the …

An Investigation Of Students' Modes Of Thinking Concerning Linearity In Linear Algebra, Noa Levy

Honors Undergraduate Theses

The intent of this thesis is to investigate student approaches to linearity within a linear algebra context, focusing on definitional, computational, and theoretical skills. Linear algebra’s abstract nature constitutes a major challenge for a significant sector of STEM students, with the course often serving as undergraduates’ first encounter with mathematical proofs and extrapolations. The current student struggle is reflected through the prominent gap in knowledge derived from a lack of a concrete understanding of rudimentary concepts (like linearity), pivotal to student success. As such, this investigation aimed to bridge this gap by considering students’ modes of thinking regarding the elementary …

Critical Convergence: Mapping The Boundaries Of How Faculty Interrogate Whiteness In The Geoscience Educational Landscape, James E. Hobbs

Educational Leadership & Policy Studies Dissertations

This study examined the role of faculty members in interrogating whiteness within geoscience education. The dominant reliance on whiteness as the primary way of knowing in geoscience education has long perpetuated a singular perspective that serves as a mechanism for reinforcing existing power structures rooted in white supremacy. Drawing on tenets from Critical whiteness Studies, Curriculum Theory, and Transformative Learning Theory, this research investigated U.S. higher education faculty members' strategies and challenges in disrupting whiteness within the geoscience curriculum.

Through critical qualitative narrative inquiry, data were collected through semi-structured interviews with geoscience faculty members across multiple institutions across the United …

Tracking The Reaction Networks Of Criegee Intermediates Formed In The Ozone-Assisted Oxidation Reactions Of C5 Acyclic And Endocyclic Alkenes, Arden M. Floyd

Honors Undergraduate Theses

Uni- and bi-molecular reactions involving Criegee intermediates (CIs), characteristic to alkene-ozone reactions, play an important role in synthetic organic, atmospheric, and combustion chemistry. However, difficulty in detecting key intermediates has prevented a thorough understanding of the details of the ozonolysis mechanism. To gain new insights into how alkene structure affects the CI network, the ozone assisted oxidation reactions of trans-2-pentene and cyclopentene are investigated between 320 K and 760 K in an atmospheric pressure jet stirred reactor (JSR). Molecular-beam mass spectrometry in conjunction with single photon tunable synchrotron VUV radiation is used to identify elusive intermediates by means of …

A Novel K-Nearest Neighbors Method Based On Generalized Feature Optimization For Precipitation Forecasting, Sean Guidry Stanteen

Mathematics Dissertations

This study introduces a novel k-nearest neighbors (kNN) method of forecasting precipitation at weather-observing stations. The method identifies numerous monthly temporal patterns to produce precipitation forecasts for a specific month. Compared to climatological forecasts, which average the observed precipitation over the prior thirty years, and other existing contemporary iterations of kNN, the proposed novel kNN method produces more accurate forecasts on a consistent basis. Specifically, the novel kNN method produces improved root mean square errors (RMSE), mean relative errors, and Nash-Sutcliffe coefficients when compared to climatological and other kNN forecasts at five weather …

Point Modules And Line Modules Of Certain Quadratic Quantum Projective Spaces, Jose E. Lozano

Mathematics Dissertations

During the past 36 years, some research in noncommutative algebra has been driven by attempts to classify AS-regular algebras of global dimension four. Such algebras are often considered to be noncommutative analogues of polynomial rings. In the 1980s, Artin, Tate, and Van den Bergh introduced a projective scheme that parametrizes the point modules over a graded algebra generated by elements of degree one. In 2002, Shelton and Vancliff introduced the concept of line scheme, which is a projective scheme that parametrizes line modules.

This dissertation is in two parts. In the first part, we consider a 1-parameter family of quadratic …

Exploring The Role Of Undergraduate And Graduate Real Analysis Experiences In The Mathematical Trajectories Of Women Mathematicians From Historically Disenfranchised Groups, Te'a Riley

Mathematics Dissertations

This phenomenological study examines the role of undergraduate and graduate Real Analysis courses in shaping the mathematical trajectories of seven women Ph.D. mathematicians from groups historically disenfranchised in mathematics.Qualitative analysis of interviews explores various aspects of their development as mathematicians with a focus on their experiences in Real Analysis. This study applies Ryan & Deci’s (1985) Self-Determination Theory's Basic Psychological Need Theory and Critical Race Theory to analyze the trajectories of the participants. The research explores how the fulfillment of basic psychological needs in their Real Analysis courses may have influenced their academic and professional journeys. The basic psychological need …

Calculus Students’ Problem-Solving Strategies On Related Rates Of Change Problems Appearing In Online Versus Paper-And-Pencil Format, Tyson Bailey

Mathematics Dissertations

This study explores first-semester calculus students’ use of mathematical problem-solving strategies while working related rates of change problems in both an online homework format and a traditional pencil-paper format. We address two research questions: (1) How do students’ mathematical problem-solving strategies when working online homework on related rates of change problems compare with their problem-solving strategies when working paper-and-pencil homework related rates of change problems? (2) What influence does the ‘view an example’ feature in online homework have on a student’s problem-solving strategies when working an online RRC homework problem? Using scores on free-response midterm exam problems on related rates …

Batch Culture Models Of The Murine Gut Microbiome & The Impact Of Simple Dormancy On Dormancy-Capable Microorganisms Models, Ana C. Mendez

Mathematics Dissertations

The proposed mathematical biology research utilizes mathematical models to gain insight into biological systems. These systems of ordinary differential equations model diverse topics, ranging from gut microbiomes to harmful algal blooms. A complete stability analysis, supporting phase plane portraits, bifurcation diagrams, and numerical simulations will accompany the models presented. In Chapter 2, the murine gut microbiome is modeled to match laboratory experiments in the literature. In these experiments, mice eat plasmid-carrying “donor” bacteria and naturally carry plasmid-free “resident” bacteria in their gut. The models aim to capture the behavior of plasmids, donor bacteria, and resident bacteria. Chapter 3 explores dormancy …

The Direct And Inverse Scattering Problems For The Third-Order Operator, Ivan Toledo

Mathematics Dissertations

We consider the full-line direct and inverse scattering problems for the third-order ordinary differential equation containing two potentials decaying sufficiently fast at infinity. The direct scattering problem consists of the determination of the scattering data set when the two potentials are known. The scattering data set is made up of the corresponding scattering coefficients and the bound-state information. On the other hand, the inverse scattering problem involves the recovery of the two potentials when the scattering data set is available. We formulate the inverse scattering problem via a related Riemann--Hilbert problem on the complex plane. We describe the recovery of …

New Methods In Electrical Source Imaging Based On Eeg And Post-Mortem Pathology Data, Julio Cesar Enciso Alva

Mathematics Dissertations

A central task for Neuroscience is to determine the location of electrical activity of neural origin inside the brain. Electrical signals can be recorded at a high resolution in time but low resolution in space, thus making it difficult to locate their source unambiguously. Electrical Source Imaging (ESI) is a particular framework for neural electrical source location; it is possible by modeling any additional information we may have about the electrical sources. For instance, minimal-norm estimators assume that the most plausible estimation is that with a lower norm. However, these estimators possess a low resolution in space.

In this work, …

A New Mechanistic Model Of Brain Metabolism With Optimal Parametrization, Alice Lubbe

Mathematics Dissertations

Models of glucose metabolism in the brain often focus on chemical exchanges and reactions that occur as part of the tricarboxylic acid cycle (TCA cycle). Experiments involving nuclear magnetic resonance (NMR) spectroscopy to detect and measure carbon-labeled isotopomers of metabolites such as glutamate in vivo inform kinetic, mechanistic models used to study metabolic pathways. In the present work, a new model with two compartments, astrocytic and neuronal, is developed using known biochemical processes and fit to experimental data coming from fully labeled glucose infusions. A gradient descent method is introduced and employed to obtain optimal flux parameter values involved in …

Towards Energy-Efficient Edge Computing For Tiny Ai Applications, Vamsi Krishna Bhagavathula

Theses and Dissertations

As artificial intelligence (AI) applications become more common on the edge of networks, like Raspberry Pi servers, it is crucial to optimize their energy use. This research project investigates how AI algorithms affect energy efficiency and resource usage on Raspberry Pi servers. Two models were created: one predicts resource usage, and the other predicts power consumption of AI algorithms on Raspberry Pi. Several factors are considered like CPU and memory use, algorithm speed, dataset size, and types of algorithms and datasets. Using regression-based methods, we model how these factors affect energy use. By converting categorical factors into numerical ones, we …

Improving The Robustness Of Neural Networks To Adversarial Patch Attacks Using Masking And Attribution Analysis, Atandra Mahalder

Honors Undergraduate Theses

Computer vision algorithms, including image classifiers and object detectors, play a pivotal role in various cyber-physical systems, spanning from facial recognition to self-driving vehicles and security surveillance. However, the emergence of real-world adversarial patches, which can be as simple as stickers, poses a significant threat to the reliability of AI models utilized within these systems. To address this challenge, several defense mechanisms such as PatchGuard, Minority Report, and (De)Randomized Smoothing have been proposed to enhance the resilience of AI models against such attacks. In this thesis, we introduce a novel framework that integrates masking with attribution analysis to robustify AI …

Estimating Modeling Parameters For Covid-19 Spread On Campus, Aviel S. Crigger

Honors Undergraduate Theses

Understanding the true burden of community transmission of communicable diseases like COVID-19 is crucial for effective public health response. Clinical cases, while important, only represent a fraction of the actual disease prevalence within a population. In this thesis, we investigate methods to estimate parameters that link clinical cases to the true disease prevalence using a modified compartmental model known as SICR (Susceptible, Infected, Cases, Recovered). We employ Bayesian inference and ensemble Markov chain Monte Carlo (MCMC) simulations to analyze clinical case data provided by the University of Central Florida Health Center from 2020 to 2022. Our goal is to estimate …

Shifting Educational Paradigms: An Analysis Of Evolving Student Expectations Within A Reformed Physical Chemistry Curriculum, Kevin E. Morales

Honors Undergraduate Theses

Students often hold negative expectations when enrolling in rigorous undergraduate courses. PhysicalChemistry is a challenging upper-level course for Chemistry majors; the course has a reputation as a demandingcourse that leads to students building negative expectations. This worsened when ambiguity existed in an instructor’s pedagogy. Not addressing those initial expectations of the students abates their motivation, which does not put them in the best place to learn, leading to poor academic performance. The course syllabus is often the first interaction a student has with the course; it can be a critical tool in (re)setting students' initial expectations and allows the instructor …

Drought Conditioning Conifer Seedlings For Post-Fire Management And Climate Change Adaptation, Anna C. Ross

Graduate Student Theses, Dissertations, & Professional Papers

Increased frequency, severity, and duration of drought and increased wildfire severity are together impacting forests of the western U.S. To encourage forest ecosystem resilience to these changing disturbance regimes, there is a need for novel silvicultural techniques that develop seedlings capable of establishing in hotter and drier climates. We evaluated the morphological and physiological effects of a drought conditioning treatment applied in a greenhouse on second- year ponderosa pine (Pinus ponderosa), western white pine (Pinus monticola), and western larch (Larix occidentalis) seedlings. Treatments included a well-watered control (75% of container capacity) and a water-limited …

Assessing Stormwater Management Pond Water Quality, Function, And The Potential Biotic Effects To Receiving Waters, Mitchell Elstone

Theses and Dissertations (Comprehensive)

The use of stormwater management ponds (SWMPs) has been increasing over the past five decades. However, an in-depth understanding of the daily performance of SWMPs and functionality during cold periods is limited. This is in part because mandated monitoring is relatively infrequent, and the assumption that SWMPs are inactive between storm events and during the winter. The goals of this research were to better understand daily stormwater (SW) characteristics, the performance of SWMPs based on current forms of evaluation and assess the potential for SWMP effluent to impact downstream biota. Influent and effluent samples from two SWMPs were collected daily …

A Computational Journey Through Conspiracy Theories: A Genealogical Approach, Mohsen Ghasemizade

Graduate College Dissertations and Theses

In an era where misinformation and conspiracy theories (CTs) proliferate, this study presents an approach to understanding and categorizing CTs through the development of a detailed `family tree'. By adopting different definitions, we explore CTs as efforts to explain events through the lens of hidden, malevolent forces, distinguishing between actual conspiracies and theoretical beliefs without empirical proof. Leveraging an analysis of 1769 articles from fact-checking websites and employing Keyphrase Extraction, we compiled a dataset that led to the identification of 769 unique conspiracies. A RoBERTa-based binary classifier was developed, achieving an F1 score of 87\%, to distinguish CTs from non-CT …

Nesting Habitat Selection And Management Of Three Freshwater Turtle Species Along The Shorelines Of Lake Champlain, Destini Acosta

Graduate College Dissertations and Theses

Habitat loss, caused by factors like urbanization and land conversion, disrupts ecosystems and can lead to population declines and even extinctions of species that rely on those habitats for survival. Loss of shoreline habitat has been widespread and reduces nesting grounds for species like freshwater turtles, often threatening their reproductive success and population viability. In the Lake Champlain Basin, several turtle species have experienced declines and recovery efforts are limited by a lack of information on habitats selected during the crucial period of nesting. This study aimed to identify and characterize nesting habitat selection of freshwater turtles and included two …

Capacitive Voltage Converter Method For The Search Of Water Ice On Celestial Bodies, Nathan Raymond

Graduate College Dissertations and Theses

Electrical Capacitance Tomography (ECT) is an advanced imaging technique used to map the permittivity of dielectric materials. ECT has diverse uses across various engineering fields, including industrial pipelines, the oil and gas sector, chemical manufacturing, and process monitoring, particularly for analyzing fluid flows. ECT can also be applied as a form of dielectric spectroscopy by measuring the response of materials under test (MUT) to changes in frequency, allowing for a detailed study of permittivity variations in a medium.One of NASA's Artemis missions is to locate and utilize in-situ resources. Transporting water to space is prohibitively expensive, with costs of $20,000 …

Building A Kinder Super Highway: Online Group Behavior Driven By Platform Design And Social Policy, Milo Trujillo

Graduate College Dissertations and Theses

Much of human socialization occurs online, and is mediated by telecommunicationsplatforms, particularly social media. These platforms both facilitate and restrict interaction in two ways: first, through the technical affordances they offer, such as conversation trees or direct messages or community self-moderation and voting; and second, through social policy, particularly regarding what content is permissible on a platform and how infractions are penalized. My work engages with platform influence over group social behavior through a series of case studies and through introducing purpose-built methodology.

I begin by examining the influence GitHub exhibits over open-source software de-velopment by contrasting the development practices …

Coalescence: A Carnivore Coexistence Curriculum That Braids Indigenous & Western Ecological Knowledge Into A Relevant And Experiential Learning Opportunity For Youth, Stephanie Anne Barron

Graduate Student Theses, Dissertations, & Professional Papers

As grizzly bears (Ursus arctos horriblis) begin to reoccupy more of their historic range, and as humans and large carnivore populations continue to increase, incidences of human carnivore conflict are on the rise. A decolonial curriculum designed in collaboration with the Confederated Salish and Kootenai Tribe’s wildlife biologists stands to increase awareness of Indigenous ecological knowledge and teach youth about the importance of coexistence with carnivores. Additionally, this project could greatly influence youth perceptions of grizzly bears and other large carnivores. This research project examines the development and implementation of a carnivore coexistence curriculum for youth that is guided by …