Treasures @ UT Dallas
Not a member yet
7697 research outputs found
Sort by
An Innovative Vibroseis Experiment to Detect the Moho Below the Valles Caldera, New Mexico
No full textIn this study, we present a new 1-D velocity-structure model that reveals several geological
structures beneath the Valles Caldera. The result is obtained by applying a sequence of processing
techniques that mainly include extended cross-correlation, NMO correction, and horizontal super-
gather stacking on Vibroseis (P-wave reflection) data collected from Valles Grande. For validating
and evaluating the robustness of the processed result, we implement and test the model by
iteratively forwarding modeling to a suite of travel time observed in processed data. These lead to
a quantitative constrained 1-D model that resolves three shallow caldera structures at depths of 1.2
km, 2.1 km, and 3 km, the possible upper and lower bound of the magma reservoir at the middle
crust with a depth of ~ 5 km and ~ 10 km respectively, and the possible Moho at near a depth of
34 km. We also found a good match of multiples that relates to the detected geological features by
incorporating and comparing the modeled data with the observed real data. In our result, the depth
of the Moho beneath the Valles Caldera is shallower compared to those results obtained from the
teleseismic data with surface waves, possibly related to the uprising materials from the mantle that
uplifts the Moho to a shallower depth as part of the Rio Grande Rift zone. This is an implication
that the Rio Grande rifting process is possibly driven by upwelling mantle flow beneath this region
Automated Commonsense Reasoning Techniques for Concurrent Data Structure Synthesis
No full textMultiprocessors are ubiquitous in today’s computing devices ranging from smartphones, IoT
micro-controllers, personal laptops to servers in data centers. Programming multiprocessors
is a non-trivial task and is largely practised as an art. Due to limits imposed by Moore’s
law on the computing power of microprocessors, writing software that runs on multiple
processors is an increasing need. Programs written to run on multiple processors are termed
concurrent programs as they contain concurrent threads of program execution over memory
that is shared. The area of concurrent data structures focuses on developing data structures
that can store information correctly when multiple threads try to access and modify the data
structure concurrently. The transition from data structures that are designed to work with
a single thread of execution to a concurrent version is non-trivial. There are several possible
ways in which threads can interfere that can result in incorrect information being stored in
the data structure. Furthermore, concurrent programs are notoriously hard to design and
debug, even when the number of lines in the code is small. At the same time, human experts
cleverly reason about all possible concurrent interactions of sequential code and arrive at
correct concurrent algorithms. This thesis captures the techniques used by a human expert
when designing concurrent data structures and makes that expert knowledge executable on a
computer. Thus, the tool that we have developed named Locksynth automatically generates
correct concurrent data structure code in a manner similar to how a human would approach
the problem. The generated code is guaranteed to be correct by construction. Automatically
generating code that is correct by construction is an important problem in computing due
to the sheer complexity of software systems being developed today. This thesis, thus, makes
research contributions to the area of automated concurrent data structure synthesis
Deep Convolutional Neural Network Encoding of Face Shape and Reflectance in Synthetic Face Images
No full textDeep Convolutional Neural Networks (DCNNs) trained for face identification recognize faces
across a wide range of imaging and appearance variations including illumination, viewpoint,
and expression. In the first part of this dissertation, I showed that identity-trained DCNNs
retain non-identity information in their top-level face representations, and that this informa-
tion is hierarchically organized in this representation (Hill et al., 2019). Specifically, the sim-
ilarity space was separated into two large clusters by gender, identities formed sub-clusters
within gender, illumination conditions clustered within identity, and viewpoints clustered
within illumination conditions. In the second part of this dissertation, I further examined
the representations generated by face identification DCNNs by separating face identity into
its constituent signals of “shape” and “reflectance”. Object classification DCNNs demon-
strate a bias for “texture” over “shape” information, whereas humans show the opposite
bias (Geirhos et al., 2018). No studies comparing “shape” and “texture” information have
yet been performed on DCNNs trained for face identification. Here, I used a 3D Morphable
Model (3DMM, Li, Bolkart, Black, Li, and Romero 2017) to determine the extent to which
face identification DCNNs encode the shape and/or spectral reflectance information in a
face. I also investigated the presence of illumination, expression, and viewpoint information
in the top-level representations of face images generated by DCNNs. Synthetic face stimuli
were generated using a 3DMM with separate components for a face shape’s “identity” and
“facial expression”, as well as spectral reflectance information in the form of a “texture map”.
The dataset comprised ten randomized levels each of face shape, reflectance, and expression,
with three levels of illumination (spotlight, ambient, 3 point), three levels of viewpoint pitch
(-30°, 0°, 30°), and five levels of viewpoint yaw (0°, 15°, 30°, 45°, 60°) in a complete factorial
design for a total of 45,000 images. All analyses were conducted with an Inception ResNet
V1-based network (Szegedy, Ioffe, Vanhoucke, & Alemi, 2017) trained on the VGGFace2
dataset (Cao, Shen, Xie, Parkhi, & Zisserman, 2018) and replicated with a ResNet-101-
based network (He, Zhang, Ren, & Sun, 2016) trained on University of Maryland’s Universe
dataset (Bansal, Castillo, Ranjan, & Chellappa, 2017; Bansal, Nanduri, Castillo, Ranjan, &
Chellappa, 2017; Guo, Zhang, Hu, He, & Gao, 2016). Area Under the Receiver Operating
Characteristic Curve (AUC) was used as a measure of information for each variable in the
top-level representation and t-distributed Stochastic Neighbor Embedding (Van der Maaten
& Hinton, 2008) was used to visualize the similarity space of top-level representations. The
results showed that both shape and reflectance information were encoded in the top-level
representation, and both signals were required for optimal performance. Shape-reflectance
bias was mediated by illumination such that the network showed a reflectance bias in ambient
and 3 point (photography style) illumination environments, whereas no bias was found under
spotlight illumination. Consistent with Hill et al. (2019), we found information about all
non-identity variables (illumination, expression, pitch, yaw) in the top-level representation,
although each of these signals was weakly encoded
Tracking the Biochemistry of Cancer Cells and Dynamics of Physical Systems Using Nuclear Magnetic Resonance
No full textNuclear magnetic resonance (NMR) spectroscopy, providing a versatile technique for analyzing cancer metabolism based on 13C NMR analysis, is one of the most important tools
for biological and more specifically cancer study purposes. The chemical shifts for 13C nuclei in organic molecules are spread out up to 200 ppm, enabling signal from each carbon
in a compound be seen as a distinct peak. The relatively weak signals obtained from the
NMR spectroscopy enables probing sensitive physical systems such as living systems without
significantly disturbing them.
In this dissertation, we have tracked 13C metabolism in different type of cancers, specially
Glioblastoma Multiforme (GBM). GBM is an aggressive type of the Central Nervous System
(CNS) tumor that grows within the brain tissue.
In this study, we have investigated how the individually used fructose and glucose sugars and
there combinations as high fructose corn syrup (HFCS) are metabolized in cultured SFxL
glioblastoma and Huh-7 hepatocellular carcinoma cells as probed by 13C NMR spectroscopy.
To understand more about cancer metabolism we did more study in glycolysis activity and
pentose phosphate pathway (PPP). We used [1,2-13C] glucose to investigate the amount of
lactate that could be produced from glycolysis versus PPP as the alternative route.
Furthermore, we have investigated the metabolism of [1,2-13C] glucose with inhibitors of
Lactate dehydrogenase A (LHDA) and sodium oxamate in GBM cells. LDHA is an important
enzyme that is active in most of tissues. LDHA catalyzes the reversible conversion of pyruvate
to lactate.
Moreover, we have investigated the spin-lattice relaxation time (T1) of water-glycerol mixtures at the earth magnetic field. The water 1H T1s at various ratios of water-glycerol
contents were measured at different temperatures ranging from 253.15 K to 353.15 K.
In summary, this PhD dissertation presents and discusses a unique tool for deciphering cancer
metabolism in vitro using NMR spectroscopy
Another Wavelength From You: Notes From Within the Algorithm
No full textAnother Wavelength From You: Notes From Within The Algorithm is a transformative art book
in the style of a 1970/80s scientific manual containing images and process documentation of
various art experiments that have been conceptually guided by artificial intelligence. Each
experiment contains elements of “intentional glitch” techniques that I have created or
manipulated using analogue technologies, coded algorithms, or a combination of both.
Individually, each piece is unique from the next, and no experimentation has been digitally
faked, e.g. no imitation through Photoshop or other similar application was performed. The book
itself, including the aesthetic treatment and experimentation within, represents art through
physical artifact — the book is the art. In addition to the scientific manual, Shades of Cool, an
11’ x 6’ hard-edge hanging sculpture made of wood, metal, and stretched fabric will be
showcased in the Edith O’Donnell Arts and Technology Building atrium at The University of
Texas at Dallas from April - July 2023. Sketches and design for this piece were digitally
rendered for Another Wavelength using artificial intelligence
Accounting for the Unaccountable: the Internal Auditor’s Role in Curbing National Corruption
No full textCorruption has been a scourge on organized society since the dawn of civilization. Social
scientists, politicians, business leaders, think tanks, and concerned citizens have all taken turns
attempting to solve the riddle of corruption. It perpetuates itself in multiple facets of daily life
including politics, governmental affairs, business, economic development, and society and
culture at large. It is also not limited by proximity constraints as corruption occurs across
multiple geographic dimensions including local, regional, and national levels. Despite societal
efforts to curb corruption, it continues to rear its ugly head with severe consequences. These
consequences pose financial and societal hardships that underpin corruption risk. This has
resulted in an increased demand for policy-based remediation efforts to curb corruption. Extant
literature focuses primarily on institutional, socioeconomic, cultural, and development factors as
determinants of corruption. However, not all countries that experience corruption do so
consistently according to these determinants. To address this gap, I examine the effect of support
for internal auditors across a sample of 180 countries. I hypothesize that countries with greater
support for internal auditors are more likely to have greater control of corruption. My research
utilizes a mixed methods approach encompassing a quantitative analysis with pooled time series
cross sectional data and a qualitative case study. The qualitative case study approach employs
survey data as well as professional and academic sources to compare the cases of the United
Arab Emirates and Saudi Arabia to ascertain determinants of corruption outcomes. Holding
economic diversification conditions constant, I argue that support for internal auditors in the
United Arab Emirates distinguishes its corruption outcome from that of its neighbor, Saudi
Arabia. Specifically, I seek to address the following research question: Why is corruption
pervasive in some countries but not others? My research contributes to the existing literature by
introducing another determinant of corruption and assessing whether that driver conveys the
variation in policy across the United Arab Emirates and Saudi Arabia
Effect of Selective Ototoxicity and Noise Exposure on the Middle Ear Muscle Reflex in Chinchilla
No full textInner and outer hair cells (IHC and OHC, respectively) are the two primary sensory cell types in
the mammalian cochlea. Despite extensive IHC innervation by auditory afferent nerve fibers
(ANF), our understanding of the role of IHC on daily listening ability is limited. Over 90% of
ANF activity arises from IHC activation; however, some auditory tasks appear unaffected by
extensive IHC loss. For example, significant IHC loss has little-to-no impact on hearing
thresholds across frequencies in the audiogram, the universal test of hearing sensitivity, but
significantly degrades tone-in-noise detection. This suggests that IHCs play a crucial role in
complex auditory tasks that are not well evaluated in routine clinical tests of hearing.
Considering the emphasis on the audiogram as a basis for treatment and rehabilitation of
individuals with hearing loss, the lack of sensitivity to IHC pathology is particularly concerning.
Previous reports have proposed that the middle ear muscle reflex (MEMR) may serve as a
sensitive diagnostic test for subclinical hearing loss. Cochlear synaptopathy (CS), believed to be
a form of subclinical hearing loss, is characterized by loss of presynaptic IHC ribbon synapses
following noise exposure or aging. CS symptoms are speculated to include normal audiometry,
poorer-than-expected hearing-in-noise ability, and an increased probability of experiencing
tinnitus or hyperacusis. Data from human studies have suggested that CS may be the underlying
pathology in individuals with normal hearing sensitivity who report disproportionately poorer
hearing-in-noise ability, and that CS and other forms of IHC pathology may be detectable using
the MEMR. The MEMR is an involuntary contraction of the middle ear stapedius muscle that
decreases the admittance of the tympanic membrane to attenuate loud acoustic signals, via a
reflex loop that is initiated by IHC transmitter release and ANF discharge. MEMR threshold and
amplitude measures can provide information regarding the integrity of a patient’s peripheral and
central auditory system. In clinical practice, MEMR threshold and amplitude are routinely
assessed as part of a standard audiological diagnostic test battery. The MEMR has recently been
suggested to be a sensitive measure of CS when measured with a wideband probe, and not the
standard clinical single-frequency probe.
Previous data has shown that MEMR thresholds measured using a 226 Hz probe tone were
unaffected by severe and selective loss of IHC in chinchillas treated with carboplatin, an
anticancer drug that destroys IHC in this species. To explore potential differences among MEMR
probes on the effect of selective IHC loss, the MEMR was measured using both a wideband
probe and a single-frequency probe in chinchillas with moderate to severe IHC loss.
To explore potential differences among MEMR probes on the effect of IHC synapse loss and
selective IHC loss, this dissertation measured responses related to the MEMR before and after
two cochlear lesion models in the chinchilla: (1) noise-induced IHC synaptopathic damage and
(2) carboplatin-induced selective IHC loss. The proposed study will gauge the sensitivity of the
MEMR to hair cell pathology and could advance our understanding of the mechanisms that
generate the MEMR
The Implications of Mandatory Corporate Social Responsibility (CSR) on Firms
No full textRecently, governments around the globe are reconsidering the role that firms play in overall
societal development. As agentic actors, many governments have implemented policies that
explicitly ask firms to share the responsibility of mitigating social, economic, and environmental
problems prevalent in society. While countries such as Australia, China, Denmark, France,
Malaysia, and South Africa now require firms to disclose their corporate social responsibility
(CSR) activities, countries including India, Indonesia, and Mauritius have mandated firms to
perform CSR activities. Such institutional transitions—including nonfinancial disclosure
requirement and mandatory CSR policies—reflect a global movement that leads us to revisit the
dominant assumption that CSR is voluntary in nature (Knudsen & Moon, 2022). Some questions
remain unaddressed: What happens when CSR is mandated? Does forcing firms to engage in
CSR activities fundamentally change the established relationships in the CSR literature—
namely, the relationships between CSR and CFP, board diversity and CSR, and CFP and CSR?
The topic of mandatory CSR has been less explored by management scholars despite its
increasing relevance. This dissertation is one of the initial attempts to fill this gap. Leveraging
the mandatory CSR policy implemented by the government of India, I use three studies to
examine the implications of mandatory CSR on firms.
The first study examines the relationship between firms’ prior CSR engagement and stock
performance around the announcement that CSR be mandated for large firms in India. I also
investigate the conditions—in particular, R&D intensity and marketing intensity—that could
mitigate the negative reaction of shareholders around the announcement. I propose the
mechanisms through which shareholders’ perceptions about CSR change if CSR is imposed on
firms. In the second study, I examine how the presence of (stigmatized) lower-caste directors on
boards influence: (1) firms’ specific CSR activities in favor of lower-caste communities, and (2)
overall CSR engagement—which benefits society at large. In addition, I investigate whether
lower-caste directors with a higher decision-making authority are more likely to overcome their
concern about stigmatized identity. Finally, in the third study, I integrate the behavioral theory of
firm and prospect theory to test the effects of consistent and inconsistent performance feedback
on CSR behavior of firms. Altogether, the findings from these studies unravel the uncharted
academic terrain of mandatory CSR—deepening and broadening understanding about CSR
Computational Simulations of DNA Associated Enzymes Using Both Quantum and Molecular Mechanics Methods
No full textThis dissertation concerns the development and usage of computational methods to explore a
variety of biological systems involved in DNA replication or repair. First, we improve the
polarization component of our group’s quantum mechanics/molecular mechanics (QM/MM)
software, Layered Interacting Chemical Models (LICHEM), by incorporating the effects of
induced dipoles from the MM into the Hamiltonian. We found that one round of induced
dipoles does not improve the interaction energy between the QM and MM regions. The
polarization catastrophe results in much higher interaction energies. Second, we explore the
effects of mutations on the β-clamp of DNA Polymerase III (E. coli). We used molecular
dynamics (MD) to simulate 6 different variants and found that there is a large reduction
in the dynamic motion of all variants as well as a change in the networks formed between
the domains of each monomer. Third, we used MD and QM/MM methods to study the
effects of 2 different mutations on the β-clamp. This work simulated both the β-clamp
and the Pol III core subunits including α, ϵ and θ. These distal mutations change the
DNA conformation found within the system altering the exonuclease reaction within the ϵ
subunit. Fourth, we used MD to determine how the drug, remdesivir, and two analogues
affect the structure of RNA-dependent RNA Polymerase (RdRp) when saturated throughout
the dsRNA. Incorporation of this drug along the RNA chain was found to have lead to an
overall destabilization of the polymerase. Additionally, remdesivir and the two analogues
studied had higher binding affinities than the natural substrate indicating their effectiveness.
Fifth, we explore the mechanism of primer synthesis in the CRISPR-Associated Primase
Polymerase from Marinitoga piezophila. Here, we used SAPT0 calculations to determine
the important components of the primer initiation complex. Last, we provide a review of
all computational studies carried out on 2 families of Iron and α–ketoglutarate–Dependent
enzymes. We also discuss new results for 2 enzymes not yet explored in the literature
Data-driven Predictive Models for Manufacturing Glass Fiber Composites and 3D-printed Metals Using Neural Networks and X-ray Imaging
No full textAdvanced manufacturing requires a close monitoring of process parameters, and real-time control
for rapid response to fine-tune the process conditions to produce high-quality products. While
multi-physics models provide high-fidelity simulation results, the computational time involved
prohibits from those to be used directly for feed-back control. The physics-based trained data-
driven models have the capabilities to replicate the multiphysics simulation results at the fast speed
required for design optimization and process control. The data-driven models take inputs such as
component geometry, process parameters, material properties, and output outcomes in the
components including temperature profile, residual stress, microstructural evolution, and material
property distribution. This investigation will focus on developing data-driven models for two
specific manufacturing processes, namely vacuum-assisted resin infusion molding (VARIM) and
wire arc additive manufacturing (WAAM). The data-driven predictive models are established
using deep machine learning (ML). Several ML models are implemented, including deep
convolutional neural network (CNN), for processing spatial information; recurrent neural network
(RNN), and long short-term memory (LSTM) for processing temporal information.
The manufacturing of large wind turbine blades requires well-controlled processing conditions to
prevent defect formation such as thermal waves. The VARIM process is the most prevalent method
implemented in the industry and is often studied and optimized using the physics-based finite
element models that provide accurate computational capabilities but suffer from high
computational costs in the meantime. Considering the limitations, an ML approach that employs a
deep CNN and RNN/LSTM model is established to predict the spatial-temporal temperature
distribution during the VARIM process. The ML model is trained with the “big data” that are
generated from the physics-based high-fidelity simulations, validated by a lab-scale VARIM
experiment conducted in the factory setting. Once fully trained, it can provide “real time”
predictions of the blade manufacturing process.
Powder-based additive manufacturing (AM) process, such as direct energy deposition (DED), is
widely used in fabricating metallic functional gradient materials (FGM) parts, which have
mechanical properties changing with locations in a part, since multiple metal powders are mixed
and used in the DED process. Hybrid manufacturing, including the DED and machining processes,
to fabricate stainless steel 316L/Inconel 718 FGM specimens are experimentally studied. The
molten pool evolution during the printing is observed; influences of the machining process on the
printed parts due to milling, including the surface roughness, and the hardness of the specimens
are evaluated. Towards the goal of sustainability and eco-friendly process for manufacturing, the
wire-feed-based AM process using WAAM provides porTable freeform fabrication capability,
along with precision manufacturing at both small- and large-scales. However, internal defects such
as porosities are often formed in additively manufactured metal components, the defects will
nucleate, grow, and coalesce to form cracks under loads, leading to eventual catastrophic failure.
To understand this failure process under loading, full-field porosity evolution in a WAAM
aluminum alloy cylinder under tension is observed with in-situ X-ray micro-computed tomography
(μCT). The analysis is performed with the assistance of a CNN algorithm that provides rapid
analysis of over thousands of slice images at various strains. The results provide quantitative
evaluations of the evolution of macropores inside the WAAM specimen under tension