Physical Sciences and Mathematics Commons^™

Representation And Strategy Learning For Variable-Size Tree Transformation Using Reinforcement Learning, Shirin Hosseini Shirvani

Computer Science and Engineering Dissertations

Trees as acyclic graphs are ubiquitous in representing different context where they encode connectivity patterns at all scales of organization, from biological systems to social networks. Trees are powerful resources which have been used many times for the exploration and discovery of interactions and properties in different context. Tree data structure representation approaches have led to remarkable discoveries in different real-world applications. In the last decades, extensive research and algorithms have been developed on tree or acyclic graph data structures with deep theoretical properties. The cost of solving these various problems ranges from simple linear time algorithms, to more complex …

Domain Adaptive Transfer Learning For Visual Classification, Ashiq Imran

Computer Science and Engineering Dissertations

Deep Neural Networks have made a significant impact on many computer vision applications with large-scale labeled datasets. However, in many applications, it is expensive and time-consuming to gather large-scale labeled data. With the limited availability of labeled data, it is challenging to obtain great performance. Moreover, in many real-world problems, transfer learning has been applied to cope with limited labeled training data. Transfer learning is a machine learning paradigm where pre-trained models on one task can be reused for another task. This dissertation investigates transfer learning and related machine learning techniques such as domain adaptation on visual categorization applications. At …

On The Efficacy Of Knowledge Graph Completion Methods, Accuracy Measures And Evaluation Protocols, Farahnaz Akrami

Computer Science and Engineering Dissertations

In the active research area of employing embedding models for knowledge graph completion, particularly for the task of link prediction, most prior studies used some specific benchmark datasets to evaluate such models. Most triples in those datasets belong to reverse and duplicate relations, which exhibit high data redundancy due to semantic duplication, correlation, or data incompleteness. This is a case of excessive data leakage—a model is trained using features that otherwise would not be available when the model needs to be applied for real prediction. There are also Cartesian product relations for which every triple formed by the Cartesian product …

Correcting Interval-Valued Expert Estimates: Empirical Formulas Explained, Laura A. Berrout Ramos, Vladik Kreinovich, Kittawit Autchariyapanitkul

Departmental Technical Reports (CS)

Experts' estimates are approximate. To make decisions based on these estimates, we need to know how accurate these estimate are. Sometimes, experts themselves estimate the accuracy of their estimates -- by providing the interval of possible values instead of a single number. In other cases, we can gauge the accuracy of the experts' estimates by asking several experts to estimates the same quantity and using the interval range of these values. In both situations, sometimes the interval is too narrow -- e.g., if an expert is overconfident. Sometimes, the interval is too wide -- if the expert is too cautious. …

Microscopic Object Classification Through Passive Motion Observations With Holographic Microscopy, Christian Lindensmith, Jay Nadeau, Manuel Bedrossian, Louis Sumrall, J. Kent Wallace, Eugene Serabyn

Physics Faculty Publications and Presentations

Digital holographic microscopy provides the ability to observe throughout a volume that is large compared to its resolution element without the need to refocus through the volume. This capability enables simultaneous observations of large numbers of small objects within such a volume. We have constructed a microscope that can observe a volume 0.4 x 0.4 x 1.0 μm with submicrometer resolution for observation of microorganisms and minerals in liquid environments in earth and on potential planetary missions. Because environmental samples are likely to contain mixtures of inorganics and microorganisms that are of comparable sizes near the resolution limit of the …

Applying Deep Learning To The Ice Cream Vendor Problem: An Extension Of The Newsvendor Problem, Gaffar Solihu

Electronic Theses and Dissertations

The Newsvendor problem is a classical supply chain problem used to develop strategies for inventory optimization. The goal of the newsvendor problem is to predict the optimal order quantity of a product to meet an uncertain demand in the future, given that the demand distribution itself is known. The Ice Cream Vendor Problem extends the classical newsvendor problem to an uncertain demand with unknown distribution, albeit a distribution that is known to depend on exogenous features. The goal is thus to estimate the order quantity that minimizes the total cost when demand does not follow any known statistical distribution. The …

Data Science And The Ice-Cream Vendor Problem, Makafui Azasoo

Electronic Theses and Dissertations

Newsvendor problems in Operations Research predict the optimal inventory levels necessary to meet uncertain demands. This thesis examines an extended version of a single period multi-product newsvendor problem known as the ice cream vendor problem. In the ice cream vendor problem, there are two products – ice cream and hot chocolate – which may be substituted for one another if the outside temperature is no too hot or not too cold. In particular, the ice cream vendor problem is a data-driven extension of the conventional newsvendor problem which does not require the assumption of a specific demand distribution, thus allowing …

Using An Inventory Of Unstable Slopes To Prioritize Probabilistic Rockfall Modeling And Acid Base Accounting In Great Smoky Mountains National Park, Thomas A. O'Shea

Electronic Theses and Dissertations

An inventory of unstable slopes along transportation corridors and performance modeling are important components of geotechnical asset management in Great Smoky Mountains National Park (GRSM). Hazards and risk were assessed for 285 unstable slopes along 151 miles of roadway. A multi-criteria model was created to select fourteen sites for two-dimensional probabilistic rockfall simulations and Acid Base Accounting (ABA) tests. Simulations indicate that rock material would likely enter the roadway at all fourteen sites. ABA test results indicate that influence of significant acid-producing potential is generally confined to slaty rocks of the Anakeesta Formation and graphitic schist of the Wehutty Formation. …

Analysis And Risk Estimation Of High Priority Unstable Rock Slopes In Great Smoky Mountains National Park, Tennessee And North Carolina, Samantha Farmer

Electronic Theses and Dissertations

Great Smoky Mountains National Park (GRSM) received 12.5 million visitors in 2020. With a high traffic volume, it is imperative roadways remain open and free from obstruction. Annual unanticipated rockfall events in GRSM often obstruct traffic flow. Using the Unstable Slope Management Program for Federal Land Management Agencies (USMP for FLMA) protocols, this study analyzes high priority unstable rock slopes through 1) creation of an unstable slope geodatabase and 2) generation of a final rockfall risk model using Co-Kriging from a preliminary risk model and susceptibility model. A secondary goal of this study is to provide risk estimation for the …

Manifold Learning With Tensorial Network Laplacians, Scott Sanders

Electronic Theses and Dissertations

The interdisciplinary field of machine learning studies algorithms in which functionality is dependent on data sets. This data is often treated as a matrix, and a variety of mathematical methods have been developed to glean information from this data structure such as matrix decomposition. The Laplacian matrix, for example, is commonly used to reconstruct networks, and the eigenpairs of this matrix are used in matrix decomposition. Moreover, concepts such as SVD matrix factorization are closely connected to manifold learning, a subfield of machine learning that assumes the observed data lie on a low-dimensional manifold embedded in a higher-dimensional space. Since …

Synthesis Of The Diazonium Zwitterionic Polymer/Monomer For Use As The Electrolyte In Polymer Electrolyte Membrane (Pem) Fuel Cells, Josiah Marshall

Electronic Theses and Dissertations

My research goals are to synthesize new zwitterionic perfluorosulfonimide (PFSI) monomer/polymers. They are expected to replace traditionally used perfluorosulfonic acid (PFSA) polymers as the electrolyte in PEM fuel cells. For the PFSI monomer preparation, we designed a nine-step synthesis route. Thus far, I have successfully completed the synthesis of 4- (2-bromotetrafluoroethoxy)-benzenesulfonyl amide, 4-acetoxybenzenesulfonic acid sodium salt, and 4-chlorosulfonyl phenyl acetate. The coupling reaction of 4-(2-bromotetrafluoroethoxy)- benzenesulfonyl amide with 4-chlorosulfonyl phenyl acetate, was troublesome due to slow reaction kinetics and byproducts. Additionally, I did a methodology study for the homopolymerziation of the perfluoro 3(oxapent-4-ene) sulfonyl fluoride monomer. We compared the weight …

Kinematics Of The Exceptionally-Short Surge Cycles Of Sít’ Kusá (Turner Glacier), Alaska, From 1983 To 2013, Andrew Nolan, William Kochtitzky, Ellyn M. Enderlin, Robert Mcnabb, Karl J. Kreutz

Geosciences Faculty Publications and Presentations

Glacier surges are periodic episodes of mass redistribution characterized by dramatic increases in ice flow velocity and, sometimes, terminus advance. We use optical satellite imagery to document five previously unexamined surge events of Sít’ Kusá (Turner Glacier) in the St. Elias Mountains of Alaska from 1983 to 2013. Surge events had an average recurrence interval of ~5 years, making it the shortest known regular recurrence interval in the world. Surge events appear to initiate in the winter, with speeds reaching up to ~25 m d⁻¹. The surges propagate down-glacier over ~2 years, resulting in maximum thinning of ~100 …

Quantitative Magnetic Resonance Imaging For The Early Prediction Of Treatment Response In Triple Negative Breast Cancer, Benjamin C. Musall

Dissertations & Theses (Open Access)

Triple Negative Breast Cancer (TNBC) is an aggressive subtype of breast cancer which lacks upregulated hormone receptors. Because of this, it is not vulnerable to clinically available targeted therapies. When treated with standard of care neoadjuvant systemic therapy (NAST), TNBC only shows approximately a 40% rate of pathologic complete response (pCR). A biomarker which could predict TNBC response to NAST early during treatment would be useful, as it would allow for non-responders to be triaged to alternative therapies and potentially allow for the treatment of responders to be de-escalated.

Quantitative Magnetic Resonance Imaging (MRI) may be used to probe and …

The Food Truck Problem, Supply Chains And Extensions Of The Newsvendor Problem, Dennis Quayesam

Electronic Theses and Dissertations

Inventory control is important to ensuring sufficient quantities of items are available tomeet demands of customers. The Newsvendor problem is a model used in Operations Research to determine optimal inventory levels for fulfilling future demands. Our study extends the newsvendor problem to a food truck problem. We used simulation to show that the food truck does not reduce to a newsvendor problem if demand depends on exogenous factors such temperature, time etc. We formulate the food truck problem as a multi-product multi-period linear program and found the dual for a single item. We use Discrete Event Simulation to solve the …

Spatial Analyses Of Gray Fossil Site Vertebrate Remains: Implications For Depositional Setting And Site Formation Processes, David Carney

Electronic Theses and Dissertations

This project uses exploratory 3D geospatial analyses to assess the taphonomy of the Gray Fossil Site (GFS). During the Pliocene, the GFS was a forested, inundated sinkhole that accumulated biological materials between 4.9-4.5 mya. This deposit contains fossils exhibiting different preservation modes: from low energy lacustrine settings to high energy colluvial deposits. All macro-paleontological materials have been mapped in situ using survey-grade instrumentation. Vertebrate skeletal material from the site is well-preserved, but the degree of skeletal articulation varies spatially within the deposit. This analysis uses geographic information systems (GIS) to analyze the distribution of mapped specimens at different spatial scales. …

Partially Oriented 6-Star Decomposition Of Some Complete Mixed Graphs, Kazeem A. Kosebinu

Electronic Theses and Dissertations

Let $M_v$ denotes a complete mixed graph on $v$ vertices, and let $S_6^i$ denotes the partial orientation of the 6-star with twice as many arcs as edges. In this work, we state and prove the necessary and sufficient conditions for the existence of $\lambda$-fold decomposition of a complete mixed graph into $S_6^i$ for $i\in\{1,2,3,4\}$. We used the difference method for our proof in some cases. We also give some general sufficient conditions for the existence of $S_6^i$-decomposition of the complete bipartite mixed graph for $i\in\{1,2,3,4\}$. Finally, this work introduces the decomposition of a complete mixed graph with a hole into …

Nitrogen-Doped Carbon Fiber Ultramicroelectrodes As Electrochemical Sensors For Detection Of Hydrogen Peroxide, Eric Wornyo

Electronic Theses and Dissertations

Carbon fiber ultramicroelectrodes (CF-UMEs) are commonly used as electrochemical probes and sensors due to their small size, fast response, and high signal-to-noise ratio. Surface modification strategies are often employed on CF-UMEs to improve their selectivity and sensitivity for desired applications. However, many modification methods are cumbersome and require expensive equipment. In this study, a simple approach known as soft nitriding is used to prepare nitrogen-doped CF-UMEs (N-CF-UMEs). Nitrogen groups introduced via soft nitriding act as electrocatalytic sites for the breakage of O-O bonds during the reduction of peroxides like H₂O₂, a common target of biosensing strategies. …

An Electrochemical Immunoassay System For Measuring Circulating Protein Biomarkers Of Pediatric Soft Tissue Sarcoma, Ivy Antwi

Electronic Theses and Dissertations

Measurement of circulating protein biomarkers associated with disease can facilitate early detection, help guide treatment strategies and improve patient outcomes beyond current standards of care. The combination of inexpensive 3D-printed flow cells and electrochemical biosensors has recently emerged as a viable platform for low-cost, reliable biomarker measurements. Here, we report an electrochemical immunoassay system based on simple graphite electrode arrays, 3D-printed flow cells, and signal-generating magnetic bead bioconjugates for simultaneous detection of three biomarker proteins (cancer antigen 125 (CA-125), midkine (MK) and osteopontin (OPN)) associated with pediatric soft tissue sarcomas (PSTS). Magnetic bead bioconjugates are functionalized with large amounts of …

Participatory Gis Mapping Highlights Indirect Use And Existence Values Of Coastal Resources And Marine Conservation Areas, Kaegan M. Scully‐Engelmeyer, Elise F. Granek, Max Nielsen-Pincus, Greg Brown

Environmental Science and Management Faculty Publications and Presentations

Consideration of social and cultural dimensions in coastal and marine planning has increased and ecosystem services provide important framing to investigate values and priorities associated with these systems. Research efforts in coastal communities offer insights on social dimensions of ocean and coastal management decisions, but questions remain about how demographics and geographic residence affect perceptions of marine resources and management. We conducted and analyzed a public participation geographic information system (PPGIS) mapping survey of Oregon residents to capture uses and perceived values of coastal and marine areas. We measured coastal values, explored regional differences in those values, and identified a …

The Foundations Of Inference And Its Application To Fundamental Physics, Nicholas Matthew Carrara

Legacy Theses & Dissertations (2009 - 2024)

This thesis concerns the foundations of inference – probability theory,entropic inference, information geometry, etc. – and its application to the Entropic Dynamics (ED) approach to Quantum Mechanics (QM) [21, 22, 41, 53, 56–61, 150–153, 165, 195, 196, 268]. The first half of this thesis, chapters 2-6, concern the development of the inference framework. We begin in chapter 2 by discussing de- ductive inference, which involves formal logic and it’s role in access- ing the truth of propositions. We eventually discover that deductive inference is incomplete, in that it can’t address situations in which we have incomplete information. This necessitates a …

Understanding Complex Human Activities In Videos : The Study Of Concurrent Activity Detection And Group Activity Recognition, Yi Wei

Legacy Theses & Dissertations (2009 - 2024)

Human activity understanding, as one of the most important task in video analysis, has been studied for decades. Great efforts have been made to push the activity recognition models towards effective and efficient representation learning. However, it is difficult to define an explicit semantic organization of activities, even for human. Current activity recognition benchmarks only organize the activity labels with shallow hierarchies, which hinders the development of activity recognition system.

Impact Of Inconsistent Imputation Models In Mediation Analysis, Bo Ye

Legacy Theses & Dissertations (2009 - 2024)

In this dissertation, we study the impact of inconsistent imputation methods in mediation analysis and its application. We present the study in three papers.

The Effect Of Differential Friction On Tropical Cyclone Development Close To Land, Minghao Zhou

Legacy Theses & Dissertations (2009 - 2024)

Hurricane Karl (2010) underwent unexpected rapid intensification in the Bay of Campeche while remaining in close proximity (within 150 km) to land. It is hypothesized that the frictionally-enhanced inflow contributed positively to this evolution, despite reduced surface fluxes and increased momentum loss over land. In this study, the role of differential friction in the development of a tropical cyclone (TC) tracking parallel to a coastline is examined based on this prototype.A series of high-resolution, semi-idealized numerical simulations with full physics were conducted on a β-plane with curvature effects of the Earth. After initializing a TC from a synoptic-scale thermal perturbation, …

Molecular Simulation Of Rna Conformational Dynamics : An Example Of Micro-Rna Targeting Messenger Rna : Mir-34a-Msirt1, Parisa Ebrahimi

Legacy Theses & Dissertations (2009 - 2024)

MicroRNA (miRNA), as a distinct class of biological regulators and a ”guide” member of non-coding RNA-protein complexes (RNPs), regulates more than 60% of protein-coding genes expression through base-pairing with targeted messenger RNA (mRNA) in the RNA-Induced Silencing Complex (RISC). Most of miRNAs identified in human, are conserved in other animals, which have preferentially conserved interaction sites particularly in 3’ untranslated regions (3’UTRs) of many human messenger mRNAs.The capability of a single miRNA to target more than hundreds of mRNAs, suggests that miRNAs influence essentially all developmental process and diseases, which also makes them interesting candidates as therapeutics agents. The primary …

Molecular Simulation Of Rna Conformational Dynamics : An Example Of Micro-Rna Targeting Messenger Rna : Mir-34a-Msirt1, Parisa Ebrahimi

Legacy Theses & Dissertations (2009 - 2024)

MicroRNA (miRNA), as a distinct class of biological regulators and a ”guide” member of non-coding RNA-protein complexes (RNPs), regulates more than 60% of protein-coding genes expression through base-pairing with targeted messenger RNA (mRNA) in the RNA-Induced Silencing Complex (RISC). Most of miRNAs identified in human, are conserved in other animals, which have preferentially conserved interaction sites particularly in 3’ untranslated regions (3’UTRs) of many human messenger mRNAs.The capability of a single miRNA to target more than hundreds of mRNAs, suggests that miRNAs influence essentially all developmental process and diseases, which also makes them interesting candidates as therapeutics agents. The primary …

Exploring Environmental And Methodological Sensitivities Of Forecasted And Observed Surface Winds And Gusts Using Underutilized Datasets, Alex Roslyn Gallagher

Legacy Theses & Dissertations (2009 - 2024)

Accurate forecasts and reliable observations of surface mean winds and gusts inhabit a vast and essential role in meteorological applications that range from wind energy, atmospheric transport, fire weather, and hazard assessment. This dissertation aims to address and explore known shortfalls in the prediction of mean winds and gusts as well as enhance evaluation and understanding of observed surface wind measurements. A variety of forecasts, methodologies, and observational dataset, many of which have been previously un- or under-utilized, are leveraged to tackle the differing needs of assessing mean winds, gusts, and the surface environment.Detailed verifications of HRRR (version 3 and …

Conditional And Marginal Imputation Models For Multilevel Data, Gang Liu

Legacy Theses & Dissertations (2009 - 2024)

This dissertation study extends sequential hierarchical regression imputation (SHRIMP) methods to multilevel datasets with three levels of nesting and proposes a marginal method based on marginalized multilevel model (MMM) framework. Specifically, the proposed model consists of two levels such that the first level relates the marginal mean of responses with covariates through a generalized regression model and the second level includes subject specific random effects within the same generalized regression model. To draw the inference on the population-averaged or subject-specified coefficients, the hierarchical regression and/or MMM is applied as the imputation and estimation models. We employ Markov Chain Monte Carlo …

Two Dimensional Nanoparticles/Nanozymes For Biosensing Augmented By Machine Learning, Nidhi Chintan Nandu

Legacy Theses & Dissertations (2009 - 2024)

Biosensing is an ever-evolving field with many resources devoted towards new approaches in sensing and refining the existing methods. For a long time, such approaches involved use of state-of-the-art instruments and were often proved to be time consuming and expensive. Nanoparticles with their unique properties, inexpensive synthesis and ease of manipulation have found purpose in myriad fields including biosensing. Different nanoparticles based on their composition, morphology, dimensionality, and surface modifications have already gained foot hold in the biosensor ranks. The unique opto-electric properties due to the quantum effect in the nanoparticles have also made them one of the leading alternatives …

It Takes Two To Tango : Understanding The Processes That Lead To Simultaneous Changes In Tropical Cyclone Intensity And Size And Communicating The Associated Hazards To Emergency Managers, Emily A. Paltz

Legacy Theses & Dissertations (2009 - 2024)

The severity of tropical cyclone (TC) hazards is modulated by both TC intensity and size. More intense TCs produce stronger storm surges and increase wind damage. Larger TCs potentially impact more people, increase the duration of TC hazards, produce stronger storm surges and increase the amount of rainfall and flooding. Thus, accurately forecasting both TC intensity and size and effectively communicating those forecasts are critical to properly preparing communities for TC impacts. Forecast accuracy can be improved by enhancing our understanding about the processes that cause changes in TC intensity and size. This research divides the Extended Best Track dataset …

Predictability Issues Associated With Near-Freezing Precipitation Type In Complex Terrain, Matthew Seymour

Legacy Theses & Dissertations (2009 - 2024)

Predictability challenges are heightened in winter weather forecasting when the environment for high-impact weather is marginal or varies over short distances. High-resolution numerical weather prediction (NWP), such as the High-Resolution Rapid Refresh (HRRR), and ensemble forecast systems, such as the Global Ensemble Forecast System (GEFS), are useful for constraining forecasts. However, their use can be challenging in marginal, near-freezing, situations when precipitation type is uncertain. Uncertainties in planetary boundary layer (PBL) and microphysics (MP) parameterizations and subtle synoptic-scale model errors brought on by differences in initial and lateral boundary conditions (IC/BCs) complicate the p-type forecast. Further, complex terrain, such as …