Physical Sciences and Mathematics Commons^™

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 …

Nash Blowups Of Toric Varieties In Prime Characteristic, Daniel Duarte, Jack Jeffries, Luis Núñez-Betancourt

Department of Mathematics: Faculty Publications

We initiate the study of the resolution of singularities properties of Nash blowups over fields of prime characteristic. We prove that the iteration of normalized Nash blowups desingularizes normal toric surfaces. We also introduce a prime characteristic version of the logarithmic Jacobian ideal of a toric variety and prove that its blowup coincides with the Nash blowup of the variety. As a consequence, the Nash blowup of a, not necessarily normal, toric variety of arbitrary dimension in prime characteristic can be described combinatorially.

Finite Element Methods For Elliptic Optimal Control Problems With General Tracking, Seonghee Jeong

LSU Doctoral Dissertations

This dissertation concerns a linear-quadratic elliptic distributed optimal control problem with pointwise state constraints in two spatial dimensions, where the cost function tracks the state at points, curves and regions of a domain.

First we explore the elliptic optimal control problem subject to pointwise control constraints. This problem is reduced into a problem that only involves the control. The solution of the reduced problem is characterized by a variational inequality. Then we introduce the elliptic optimal control problem with general tracking and pointwise state constraints. Here we reformulate the optimal control problem into a problem that only involves the state, …

First-Order Algorithms For Nonlinear Structured Optimization, Miao Zhang

LSU Doctoral Dissertations

Nonlinear optimization is a critical branch in applied mathematics and has attracted wide attention due to its popularity in practical applications. In this work, we present two methods which use first-order information to solve two typical classes of nonlinear structured optimization problems.

For a class of unconstrained nonconvex composite optimization problems where the objective is the sum of a smooth but possibly nonconvex function and a convex but possibly nonsmooth function, we propose a unified proximal gradient method with extrapolation, which provides unified treatment to convex and nonconvex problems. The method achieves the best-known convergence rate for first-order methods when …

Extremal Absorbing Sets In Low-Density Parity-Check Codes, Emily Mcmillon, Allison Beemer, Christine A. Kelley

Department of Mathematics: Faculty Publications

Absorbing sets are combinatorial structures in the Tanner graphs of low-density parity-check (LDPC) codes that have been shown to inhibit the high signal-to-noise ratio performance of iterative decoders over many communication channels. Absorbing sets of minimum size are the most likely to cause errors, and thus have been the focus of much research. In this paper, we determine the sizes of absorbing sets that can occur in general and left-regular LDPC code graphs, with emphasis on the range of b for a given a for which an (a, b)-absorbing set may exist. We identify certain cases of extremal …

Lagrangian Mechanics And The Motion Of A T-Handle, Eric Maurer

WWU Honors College Senior Projects

This paper begins by deriving the equations of motion for the Lagrangian formulation of mechanics. Lagrangian mechanics describes the same thing as the traditional Newtonian mechanics (which is what is taught in most undergrad physics classes), but rather than model the system through forces, it models the system through energy. Energy is conserved in a system, which allows the Lagrangian formulation to model certain types of systems in a more efficient way than the Newtonian formulation.

Next, the paper explores rigid body motion, specifically looking at the motion of the angular momentum. The goal of this part is to explain …

Solving The Genius Square: Using Math And Computers To Analyze A Polyomino Tiling Game, Noah Jensen

WWU Honors College Senior Projects

This paper investigates if the claim of the game Genius Square is true, that all 62,208 boards that its dice can roll are solvable. The gameboard is a 6 by 6 grid and the objective of the game is to tile a board that has 7 blockers, quasi-randomly placed by the dice, with 9 polyominoes consisting of 1, 2, 3, and 4 squares. In order to implement a model of linear systems created by John Burkardt and M.R. Garvie, code was developed using Python and Matlab. With this code, it was shown that all 62,208 boards are solvable. The number …

Modeling Immune System Dynamics During Hiv Infection And Treatment With Differential Equations, Nicole Rychagov

CODEE Journal

An inquiry-based project that discusses immune system dynamics during HIV infection using differential equations is presented. The complex interactions between healthy T-cells, latently infected T-cells, actively infected T-cells, and the HIV virus are modeled using four nonlinear differential equations. The model is adapted to simulate long term HIV dynamics, including the AIDS state, and is used to simulate the long term effects of the traditional antiretroviral therapy (ART). The model is also used to test viral rebound over time of combined application of ART and a new drug that blocks the reactivation of the viral genome in the infected cells …

Everyone's A Waiter: A Data-Driven Queuing Simulation Model Of Mike's Clam Shack, Natalie Robinson

Honors College

This thesis seeks to understand the mathematical foundation of several prominent concepts in queuing theory and apply them to gain a better understanding of nightly business levels and dining room queue behavior during the summer tourist season at Mike’s Clam Shack, which is a restaurant located in Wells, Maine. To do so, a variety of queue and server section data has been collected from Mike’s and analyzed to determine probability distributions for interarrival and service times. In addition, a queuing simulation model has been constructed in the R Programming Language, which uses this data to generate dining room and queue …

Practical Implementation Of The Immersed Interface Method With Triangular Meshes For 3d Rigid Solids In A Fluid Flow, Norah Hakami

Mathematics Theses and Dissertations

When employing the immersed interface method (IIM) to simulate a fluid flow around a moving rigid object, the immersed object can be replaced by a virtual fluid enclosed by singular forces on the interface between the real and virtual fluids. These forces represent the impact of the rigid motion on the fluid flow and cause jump discontinuities across the interface in the whole flow field. Then, the IIM resolves the fluid flow on a fixed computational domain by directly incorporating the jump conditions across the interface into numerical schemes. Previous development of the method is limited to simple smooth boundaries. …

Viscous Thin-Film Models Of Nanoscale Self-Organization Under Ion Bombardment, Tyler Evans

Mathematics Theses and Dissertations

For decades, it has been observed that broad-beam irradiation of semiconductor surfaces can lead to spontaneous self-organization into highly regular patterns, sometimes at length scales of only a few nanometers. Initial theory was largely based on erosion and redistribution of material occurring on fast time scales, which are able to produce good agreement with certain aspects of surface evolution. However, further experimental and theoretical work eventually led to the realization that numerous effects are active in the irradiated target, including stresses associated with ion-implantation and the accumulation of damage leading to the development of a disordered, amorphous layer atop the …

A Node Elimination Algorithm For Cubatures Of High-Dimensional Polytopes, Arkadijs Slobodkins

Mathematics Theses and Dissertations

Node elimination is a numerical approach for obtaining cubature rules for the approximation of multivariate integrals over domains in Rⁿ. Beginning with a known cubature, nodes are selected for elimination, and a new, more efficient rule is constructed by iteratively solving the moment equations. In this work, a new node elimination criterion is introduced that is based on linearization of the moment equations. In addition, a penalized iterative solver is introduced that ensures positivity of weights and interiority of nodes. We aim to construct a universal algorithm for convex polytopes that produces efficient cubature rules without any user …

Data-Driven Learning Algorithm Via Densely-Defined Multiplication Operators And Occupation Kernels., John Kyei

USF Tampa Graduate Theses and Dissertations

Consider a nonautonomous nonlinear evolution $\dot{x}=f(x,t,\mu)$, where the vector $x(t) \in \mathbb{R}^n$ represents the state of the dynamical system at time $t$, $\mu$ contains system parameters, and $f(\cdot)$ represents a dynamic constraint. In most practical applications, the nonlinear dynamic constraint $f$ is unknown analytically. The problem of approximating $f$ directly from data measurements generated by the system is a main goal of this manuscript. In the postulates of the Nonlinear Autoregressive (NAR) framework, we show that the problem of approximating $f$ can be studied through symbols of densely defined multiplication operators over a Reproducing Kernel Hilbert Spaces (RKHS). In this …

Using Physics-Informed Neural Networks For Multigrid In Time Coarse Grid Equations, Jonathan P. Gutierrez

Mathematics & Statistics ETDs

For parallel-in-time integration methods, the multigrid-reduction-in-time (MGRIT) method has shown promising results in both improved convergence and increased computational speeds when solving evolution problems. However, one problem the MGRIT algorithm currently faces is it struggles solving hyperbolic problems efficiently. In particular, hyperbolic problems are generally solved using explicit methods and this causes issues on the coarser multigrid levels, where larger (coarser) time step sizes can violate the stability condition. In this thesis, physics-informed neural networks (PINNs) are used to evaluate the coarse grid equations in the MGRIT algorithm with the goal to improve convergence for problems with hyperbolic behavior, as …

Mathematical Modeling And Inverse Problems In Applications, Thanh T. Nguyen

College of Science & Mathematics Departmental Research

Mathematical models, based on ordinary or partial differential equations, are widely used to describe physical/chemical/biological processes and can be found in several applications: nondestructive testing, subsurface imaging, defense, medicine, environmental sciences, etc.

Problem Of The Week: A Student-Led Initiative To Bring Mathematics To A Broader Audience, Jordan M. Sahs, Brad Horner

UNO Student Research and Creative Activity Fair

Problem of the Week (POW!) is a weekly undergraduate mathematics competition hosted by two graduate students from the UNO Math Department. It started with the goal to showcase variety, creativity, and intrigue in math to those who normally feel math is dry, rote, and formulaic. Problems shine light on both hidden gems and popular recreational math, both math history and contemporary research, both iconic topics and nontraditional ones, both pure abstraction and real-world application. Now POW! aims to increase availability and visibility in Omaha and beyond. Select problems from Fall 2021 to Spring 2023 are highlighted here: these received noteworthy …

Formal Conjugacy Growth In Graph Products I, Laura Ciobanu,, Susan Hermiller, Valentin Mercier

Department of Mathematics: Faculty Publications

In this paper we give a recursive formula for the conjugacy growth series of a graph product in terms of the conjugacy growth and standard growth series of subgraph products. We also show that the conjugacy and standard growth rates in a graph product are equal provided that this property holds for each vertex group. All results are obtained for the standard generating set consisting of the union of generating sets of the vertex groups.

Chatgpt As Metamorphosis Designer For The Future Of Artificial Intelligence (Ai): A Conceptual Investigation, Amarjit Kumar Singh (Library Assistant), Dr. Pankaj Mathur (Deputy Librarian)

Library Philosophy and Practice (e-journal)

Abstract

Purpose: The purpose of this research paper is to explore ChatGPT’s potential as an innovative designer tool for the future development of artificial intelligence. Specifically, this conceptual investigation aims to analyze ChatGPT’s capabilities as a tool for designing and developing near about human intelligent systems for futuristic used and developed in the field of Artificial Intelligence (AI). Also with the helps of this paper, researchers are analyzed the strengths and weaknesses of ChatGPT as a tool, and identify possible areas for improvement in its development and implementation. This investigation focused on the various features and functions of ChatGPT that …

Student Performance In Traditional In-Person Vs. Online Sections Of An Introductory Graduate Mathematics Course, Lauran E. Kittle

Theses and Dissertations

The growth of technology impacts nearly every aspect of everyday life, to include education and learning. The availability of distance learning (online) classes has increased drastically in the last few decades, expanding access to education for millions of people. However, it is imperative to consider exactly how the growth of technology impacts education – whether it is a positive, negative, or neutral impact. Previous research comparing distance learning and in-residence (traditional) classes have widely mixed, disparate conclusions. This type of research, two-stage analysis, and modeling has yet to be conducted on a graduate school level. For this reason, a detailed …

Mathematical Models For Thalassemia, Hamda Mohammed Al Dhaheri

Theses

Thalassemia is a genetic blood disorder caused by gene mutation or deletion in a blood protein called hemoglobin. Treatment of thalassemia requires a life-long blood transfusion and removal of excessive iron in the blood stream, which usually causes a big pressure on health care systems. Various forms of thalassemia control measures have been used to reduce the prevalence of thalassemia major. This has resulted in a substantial reduction in the prevalence of thalassemia. However, the thalassemia carrier population remains high, which could lead to an increase in the thalassemia major population through carrier-to-carrier marriages. Thus, we developed two mathematical models …

Evolution Of Coronal Magnetic Field Parameters During X5.4 Solar Flare, Seth H. Garland, Benjamin F. Akers, Vasyl B. Yurchyshyn, Robert D. Loper, Daniel J. Emmons

Faculty Publications

The coronal magnetic field over NOAA Active Region 11,429 during a X5.4 solar flare on 7 March 2012 is modeled using optimization based Non-Linear Force-Free Field extrapolation. Specifically, 3D magnetic fields were modeled for 11 timesteps using the 12-min cadence Solar Dynamics Observatory (SDO) Helioseismic and Magnetic Imager photospheric vector magnetic field data, spanning a time period of 1 hour before through 1 hour after the start of the flare. Using the modeled coronal magnetic field data, seven different magnetic field parameters were calculated for 3 separate regions: areas with surface |Bz| ≥ 300 G, areas of flare brightening seen …

Analysis Of The Accuracy Of The Solution Of The Integral Equation Of The Interpretation Problem, Majid Malikovich Karimov, Mirhusan Mirazizovich Sagatov

Chemical Technology, Control and Management

Intensive development and expansion of the field of application of modern observation systems are largely determined by the improvement of methods and means of interpreting the primary results recorded by the system. The high quality of the results of solving interpretation problems is achieved by improving such indicators as accuracy, resolution, speed. In the general case, the influence of the observation system on the result is naturally described by an integral equation, which is a general integral mathematical model for the problems of interpreting the results of observations. In the presented article, it is proposed to improve the accuracy of …

Bernstein-Sato Theory For Singular Rings In Positive Characteristic, Jack Jack, Luis Núñez-Betancourt, Eamon Quinlan-Gallego

Department of Mathematics: Faculty Publications

The Bernstein-Sato polynomial is an important invariant of an element or an ideal in a polynomial ring or power series ring of characteristic zero, with interesting connections to various algebraic and topological aspects of the singularities of the vanishing locus. Work of Mustaţă, later extended by Bitoun and the third author, provides an analogous Bernstein-Sato theory for regular rings of positive characteristic.

In this paper, we extend this theory to singular ambient rings in positive characteristic. We establish finiteness and rationality results for Bernstein-Sato roots for large classes of singular rings, and relate these roots to other classes of numerical …

Tri-Plane Diagrams For Simple Surfaces In S4, Manuel Aragón, Zack Dooley, Alexander Goldman, Yucong Lei, Isaiah Martinez, Nicholas Meyer, Devon Peters, Scott Warrander, Ana Wright, Alex Zupan

Department of Mathematics: Faculty Publications

Meier and Zupan proved that an orientable surface K in S⁴ admits a tri-plane diagram with zero crossings if and only if K is unknotted, so that the crossing number of K is zero. We determine the minimal crossing numbers of nonorientable unknotted surfaces in S⁴, proving that c(P^n,m) = max{1, |n−m|}, where P^n,m denotes the connected sum of n unknotted projective planes with normal Euler number +2 and m unknotted projective planes with normal Euler number −2. In addition, we convert Yoshikawa’s table of knotted surface ch-diagrams to tri-plane …

Session 12: Analysis Of State And Parameter Estimation Techniques Using Dynamic Perturbation Signals, Timothy M. Hansen

SDSU Data Science Symposium

The trend in electric power systems is the displacement of traditional synchronous generation (e.g., coal, natural gas) with renewable energy resources (e.g., wind, solar photovoltaic) and battery energy storage. These energy resources require power electronic converters (PECs) to interconnect to the grid and have different response characteristics and dynamic stability issues compared to conventional synchronous generators. As a result, there is a need for validated models to study and mitigate PEC-based stability issues, especially for converter dominated power systems (e.g., island power systems, remote microgrids).

This presentation will introduce methods related to dynamic state and parameter estimation via the design …

Decomposition Rate As An Emergent Property Of Optimal Microbial Foraging, Stefano Manzoni, Arjun Chakrawal, Glenn Ledder

Department of Mathematics: Faculty Publications

Decomposition kinetics are fundamental for quantifying carbon and nutrient cycling in terrestrial and aquatic ecosystems. Several theories have been proposed to construct process-based kinetics laws, but most of these theories do not consider that microbial decomposers can adapt to environmental conditions, thereby modulating decomposition. Starting from the assumption that a homogeneous microbial community maximizes its growth rate over the period of decomposition, we formalize decomposition as an optimal control problem where the decomposition rate is a control variable. When maintenance respiration is negligible, we find that the optimal decomposition kinetics scale as the square root of the substrate concentration, resulting …

Pillars Of Biology: 'The Genetical Evolution Of Social Behaviour, I And Ii'., Geoff Wild

Applied Mathematics Publications

None.

A Statistical Analysis Of The Change In Age Distribution Of Spawning Hatchery Salmon, Rachel Macaulay, Emily Barrett, Grace Penunuri, Eli E. Goldwyn

Spora: A Journal of Biomathematics

Declines in salmon sizes have been reported primarily as a result of younger maturation rates. This change in age distribution poses serious threats to salmon-dependent peoples and ecological systems. We perform a statistical analysis to examine the change in age structure of spawning Alaskan chum salmon Oncorhynchus keta and Chinook salmon O. tshawytscha using 30 years of hatchery data. To highlight the impacts of this change, we investigate the average number of fry/smolt that each age of spawning chum/Chinook salmon produce. Our findings demonstrate an increase in younger hatchery salmon populations returning to spawn, and fewer amounts of fry produced …

Spike-Time Neural Codes And Their Implication For Memory, Alexandra Busch

Electronic Thesis and Dissertation Repository

The possibility of temporal coding in neural data through patterns of precise spike times has long been of interest in neuroscience. Recent and rapid advancements in experimental neuroscience make it not only possible, but also routine, to record the spikes of hundreds to thousands of cells simultaneously. These increasingly common large-scale data sets provide new opportunities to discover temporally precise and behaviourally relevant patterns of spiking activity across large populations of cells. At the same time, the exponential growth in size and complexity of new data sets presents its own methodological challenges. Specifically, it remains unclear how best to (1) …

Pythagorean Vectors And Rational Orthonormal Matrices, Aishat Olagunju

Electronic Thesis and Dissertation Repository

A Pythagorean vector is an integer vector having an integer 2-norm. Such vectors are closely related to Pythagorean n-tuples, since n-tuples are the building blocks for Pythagorean vectors. Pythagorean vectors are, in their turn, the building blocks for rational orthonormal matrices. The work in this thesis has a pedagogical application to the QR decomposition of matrices, widely used in Linear Algebra. A barrier for students learning the details of the QR decomposition of a given matrix A is the occurrence of square-roots that cannot be simplified during the application of the two standard algorithms, namely the Gram--Schmidt method and Householder …