Physical Sciences and Mathematics Commons^™

Assessing Extant Methods For Generating G-Optimal Designs And A Novel Methodology To Compute The G-Score Of A Candidate Design, Hyrum John Hansen

All Graduate Theses and Dissertations, Fall 2023 to Present

Experimental designs are used by scientists to allocate treatments such that statistical inference is appropriate. Most traditional experimental designs have mathematical properties that make them desirable under certain conditions. Optimal experimental designs are those where the researcher can exercise total control over the treatment levels to maximize a chosen mathematical property. As is common in literature, the experimental design is represented as a matrix where each column represents a variable, and each row represents a trial. We define a function that takes as input the design matrix and outputs its score. We then algorithmically adjust each entry until a design …

Classification In Supervised Statistical Learning With The New Weighted Newton-Raphson Method, Toma Debnath

Electronic Theses and Dissertations

In this thesis, the Weighted Newton-Raphson Method (WNRM), an innovative optimization technique, is introduced in statistical supervised learning for categorization and applied to a diabetes predictive model, to find maximum likelihood estimates. The iterative optimization method solves nonlinear systems of equations with singular Jacobian matrices and is a modification of the ordinary Newton-Raphson algorithm. The quadratic convergence of the WNRM, and high efficiency for optimizing nonlinear likelihood functions, whenever singularity in the Jacobians occur allow for an easy inclusion to classical categorization and generalized linear models such as the Logistic Regression model in supervised learning. The WNRM is thoroughly investigated …

A Machine Learning Approach For Predicting Clinical Trial Patient Enrollment In Drug Development Portfolio Demand Planning, Ahmed Shoieb

Masters Theses

One of the biggest challenges the clinical research industry currently faces is the accurate forecasting of patient enrollment (namely if and when a clinical trial will achieve full enrollment), as the stochastic behavior of enrollment can significantly contribute to delays in the development of new drugs, increases in duration and costs of clinical trials, and the over- or under- estimation of clinical supply. This study proposes a Machine Learning model using a Fully Convolutional Network (FCN) that is trained on a dataset of 100,000 patient enrollment data points including patient age, patient gender, patient disease, investigational product, study phase, blinded …

Multilevel Optimization With Dropout For Neural Networks, Gary Joseph Saavedra

Mathematics & Statistics ETDs

Large neural networks have become ubiquitous in machine learning. Despite their widespread use, the optimization process for training a neural network remains com-putationally expensive and does not necessarily create networks that generalize well to unseen data. In addition, the difficulty of training increases as the size of the neural network grows. In this thesis, we introduce the novel MGDrop and SMGDrop algorithms which use a multigrid optimization scheme with a dropout coarsening operator to train neural networks. In contrast to other standard neural network training schemes, MGDrop explicitly utilizes information from smaller sub-networks which act as approximations of the full …

Abm Simulation Model Of A Pandemic For Optimizing Vaccination Strategy, Gibeom Park

Theses and Dissertations

This study presents a process-oriented hybrid model for individuals' immune responses and interactions involving vaccination to describe the trend of contagious disease and estimate the future societal cost. The model considers "recovery" as a non-absorbing state and incorporates various infection stage states including two symptomatic states. To model contagiousness to be consistent with the current pandemic and include that the spread of a disease depends on the mobility of people, we developed an Agent-Based Simulator that fitted to the particular model used in this study and can test various what-if scenarios. We improved the simulator considerably by appying data structures …

Debiasing Cyber Incidents – Correcting For Reporting Delays And Under-Reporting, Seema Sangari

Doctor of Data Science and Analytics Dissertations

This research addresses two key problems in the cyber insurance industry – reporting delays and under-reporting of cyber incidents. Both problems are important to understand the true picture of cyber incident rates. While reporting delays addresses the problem of delays in reporting due to delays in timely detection, under-reporting addresses the problem of cyber incidents frequently under-reported due to brand damage, reputation risk and eventual financial impacts.

The problem of reporting delays in cyber incidents is resolved by generating the distribution of reporting delays and fitting modeled parametric distributions on the given domain. The reporting delay distribution was found to …

Development Of A Reverse Engineered, Parameterized, And Structurally Validated Computational Model To Identify Design Parameters That Influence American Football Faceguard Performance, William Ferriell

All Dissertations

Traumatic brain injury (TBI) continues to have the greatest incidence among athletes participating in American football. The headgear design research community has focused on developing accurate computational and experimental analysis techniques to better assess the ability of headgear technology to attenuate impacts and protect athletes from TBI. Despite efforts to innovate the headgear system, minimal progress has been made to innovate the faceguard. Although the faceguard is not the primary component of the headgear system that contributes to impact attenuation, faceguard performance metrics, such as weight, structural stiffness, and visual field occlusions, have been linked to athlete safety. To improve …

Optimizing Critical Values And Combining Axes For Multi-Axial Neck Injury Criteria, Ethan J. Gaston

Theses and Dissertations

The Air Force employs ejection seats in its high-performance aircraft. While these systems are intended to ensure aircrew safety, the ejection process subjects the aircrew to potentially injurious forces. System validation includes evaluation of forces against a standard which is linked to the probability of injury. The Muti-Axial Neck Injury Criteria (MANIC) was developed to account for forces in all six degrees of freedom. Unfortunately, the MANIC is applied to each of the three linear input directions separately and applies different criterion values for each direction. These three separate criteria create a lack of clarity regarding acceptable neck loading, leading …

Machine Learning Morphisms: A Framework For Designing And Analyzing Machine Learning Work Ows, Applied To Separability, Error Bounds, And 30-Day Hospital Readmissions, Eric Zenon Cawi

McKelvey School of Engineering Theses & Dissertations

A machine learning workflow is the sequence of tasks necessary to implement a machine learning application, including data collection, preprocessing, feature engineering, exploratory analysis, and model training/selection. In this dissertation we propose the Machine Learning Morphism (MLM) as a mathematical framework to describe the tasks in a workflow. The MLM is a tuple consisting of: Input Space, Output Space, Learning Morphism, Parameter Prior, Empirical Risk Function. This contains the information necessary to learn the parameters of the learning morphism, which represents a workflow task. In chapter 1, we give a short review of typical tasks present in a workflow, as …

Characterizing Uncertainty In Correlated Response Variables For Pareto Front Optimization, Peter A. Calhoun

Theses and Dissertations

Current research provides a method to incorporate uncertainty into Pareto front optimization by simulating additional response surface model parameters according to a Multivariate Normal Distribution (MVN). This research shows that analogous to the univariate case, the MVN understates uncertainty, leading to overconfident conclusions when variance is not known and there are few observations (less than 25-30 per response). This research builds upon current methods using simulated response surface model parameters that are distributed according to an Multivariate t-Distribution (MVT), which can be shown to produce a more accurate inference when variance is not known. The MVT better addresses uncertainty in …

Paper Structure Formation Simulation, Tyler R. Seekins

Electronic Theses and Dissertations

On the surface, paper appears simple, but closer inspection yields a rich collection of chaotic dynamics and random variables. Predictive simulation of paper product properties is desirable for screening candidate experiments and optimizing recipes but existing models are inadequate for practical use. We present a novel structure simulation and generation system designed to narrow the gap between mathematical model and practical prediction. Realistic inputs to the system are preserved as randomly distributed variables. Rapid fiber placement (~1 second/fiber) is achieved with probabilistic approximation of chaotic fluid dynamics and minimization of potential energy to determine flexible fiber conformations. Resulting digital packed …

Generalized Clusterwise Regression For Simultaneous Estimation Of Optimal Pavement Clusters And Performance Models, Mukesh Khadka

UNLV Theses, Dissertations, Professional Papers, and Capstones

The existing state-of-the-art approach of Clusterwise Regression (CR) to estimate pavement performance models (PPMs) pre-specifies explanatory variables without testing their significance; as an input, this approach requires the number of clusters for a given data set. Time-consuming ‘trial and error’ methods are required to determine the optimal number of clusters. A common objective function is the minimization of the total sum of squared errors (SSE). Given that SSE decreases monotonically as a function of the number of clusters, the optimal number of clusters with minimum SSE always is the total number of data points. Hence, the minimization of SSE is …

Deterministic And Probabilistic Methods For Seismic Source Inversion, Juan Pablo Madrigal Cianci

Mathematics & Statistics ETDs

The national Earthquake Information Center (NEIC) reports an occurrence of about 13,000 earthquakes every year, spanning different values on the Richter scale from very mild (2) to "giant earthquakes'' (8 and above). Being able to study these earthquakes provides useful information for a wide range of applications in geophysics. In the present work we study the characteristics of an earthquake by performing seismic source inversion; a mathematical problem that, given some recorded data, produces a set of parameters that when used as input in a mathematical model for the earthquake generates synthetic data that closely resembles the measured data. There …

Inference In Networking Systems With Designed Measurements, Chang Liu

Doctoral Dissertations

Networking systems consist of network infrastructures and the end-hosts have been essential in supporting our daily communication, delivering huge amount of content and large number of services, and providing large scale distributed computing. To monitor and optimize the performance of such networking systems, or to provide flexible functionalities for the applications running on top of them, it is important to know the internal metrics of the networking systems such as link loss rates or path delays. The internal metrics are often not directly available due to the scale and complexity of the networking systems. This motivates the techniques of inference …

Probability Models For Health Care Operations With Application To Emergency Medicine, Azaz Bin Sharif

Electronic Thesis and Dissertation Repository

This thesis consists of four contributing chapters; two of which are inspired by practical problems related to emergency department (ED) operations management and the remaining two are motivated by the theoretical problem related to the time-dependent priority queue. Unlike classical priority queue, priorities in the time-dependent priority queue depends on the amount of time an arrival waits for service in addition to the priority class they belong. The mismatch between the demand for ED services and the available resources have direct and indirect negative consequences. Moreover, ED physician pay in some jurisdictions reflects pay-for-performance contracts based on operational benchmarks. To …

Recent Advances In Accumulating Priority Queues, Na Li

Electronic Thesis and Dissertation Repository

This thesis extends the theory underlying the Accumulating Priority Queue (APQ) in three directions. In the first, we present a multi-class multi-server accumulating priority queue with Poisson arrivals and heterogeneous services. The waiting time distributions for different classes have been derived. A conservation law for systems with heterogeneous servers has been studied. We also investigate an optimization problem to find the optimal level of heterogeneity in the multi-server system. Numerical investigations through simulation are carried out to validate the model.

We next focus on a queueing system with Poisson arrivals, generally distributed service times and nonlinear priority accumulation functions. We …

Developing An Optimal Model For Infant Home Visitation, Isaac Atuahene

Doctoral Dissertations

The United States, Great Britain, Denmark, Canada and many other countries have accepted home visitation (HV) as a promising strategy for interventions for infants after births and for their mothers. Prior HV studies have focused on theoretical foundations, evaluations of programs, cost/benefit analysis and cost estimation by using hospital/payer/insurance data to prove its effectiveness and high cost. As governments and private organizations continue to fund HVs, it is an opportune time to develop and formulate operations research (OR) models of HV coverage, quality and cost so they might be used in program implementation as done for adult home healthcare (HHC) …

Poisson Distributed Individuals Control Charts With Optimal Limits, Negin Enayaty Ahangar

Graduate Theses and Dissertations

The conventional method used in attribute control charts is the Shewhart three sigma limits. The implicit assumption of the Normal distribution in this approach is not appropriate for skewed distributions such as Poisson, Geometric and Negative Binomial. Normal approximations perform poorly in the tail area of the these distributions. In this research, a type of attribute control chart is introduced to monitor the processes that provide count data. The economic objective of this chart is to minimize the cost of its errors which is determined by the designer. This objective is a linear function of type I and II errors. …

Performance Modeling And Optimization Techniques For Heterogeneous Computing, Supada Laosooksathit

Doctoral Dissertations

Since Graphics Processing Units (CPUs) have increasingly gained popularity amoung non-graphic and computational applications, known as General-Purpose computation on GPU (GPGPU), CPUs have been deployed in many clusters, including the world's fastest supercomputer. However, to make the most efficiency from a GPU system, one should consider both performance and reliability of the system.

This dissertation makes four major contributions. First, the two-level checkpoint/restart protocol that aims to reduce the checkpoint and recovery costs with a latency hiding strategy in a system between a CPU (Central Processing Unit) and a GPU is proposed. The experimental results and analysis reveals some benefits, …

Optimization In Non-Parametric Survival Analysis And Climate Change Modeling, Iuliana Teodorescu

USF Tampa Graduate Theses and Dissertations

Many of the open problems of current interest in probability and statistics involve complicated data

sets that do not satisfy the strong assumptions of being independent and identically distributed. Often,

the samples are known only empirically, and making assumptions about underlying parametric

distributions is not warranted by the insufficient information available. Under such circumstances,

the usual Fisher or parametric Bayes approaches cannot be used to model the data or make predictions.

However, this situation is quite often encountered in some of the main challenges facing statistical,

data-driven studies of climate change, clinical studies, or financial markets, to name a few. …

An Integrated Screening And Optimization Strategy, Nathaniel Jackson Rohbock

Theses and Dissertations

Within statistical methods, design of experiments (DOE) is well suited to make good inference from a minimal amount of data. Two types of designs within DOE are screening designs and optimization designs. Traditionally, these approaches have been necessarily separated by a gap between the objectives of each design and the methods available. Despite being so separated, in practice these designs are frequently connected by sequential experimentation. In fact, from the genesis of a project, the experimentor often knows that both designs will be necessary to accomplish his objectives. Due to advances in the understanding of experimental designs with complex aliasing …

Computer-Based Methods For Constructing Two-Level Fractional-Factorial Experimental Designs With A Requirement Set, Steven L. Forsythe

Theses and Dissertations

This dissertation developed four methodologies for computer-aided experimental design of two-level fractional factorial designs with requirement sets (DOE/RS). The requirement sets identify all the experimental factors and the appropriate interaction terms to be evaluated in the experiment. Taguchi graphs and similar manual methods provide techniques for solving the DOE/RS problem. Unfortunately, these methods are limited because they become difficult to use as the number of factors or interaction terms exceeds ten. This research showed that the DOE/RS problem belongs to a class of difficult-to-solve problems known as NP-Complete. It is the combinatorial nature of NP-Complete problems that causes them to …

Design Optimization Using Model Estimation Programming, Richard Kay Brimhall

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Model estimation programming provides a method for obtaining extreme solutions subject to constraints. Functions which are continuous with continuous first and second derivatives in the neighborhood of the solution are approximated using quadratic polynomials (termed estimating functions) derived from computed or experimental data points. Using the estimating functions, an approximation problem is solved by a numerical adaptation of the method of Lagrange. The method is not limited by the concavity of the objective function.

Beginning with an initial array of data observations, an initial approximate solution is obtained. Using this approximate solution as a new datum point, the coefficients for …