Treasures @ UT Dallas
Not a member yet
7697 research outputs found
Sort by
The Impact of Work-life Balance Policies on Perceived Organizational Performance in the U.S. Federal Agencies: Exploring the Moderating Effect of Leadership Support
No full textThis study enhances our understanding of the implementation of work-life balance policies and
its impact on perceived organizational performance in U.S. federal agencies. While interest in
implementing work-life balance policies began in the 1960s and 1970s after the influx of females
into the labor market, few studies have examined the impact of these policies on organizational
performance in the public organizations context, particularly in the federal agencies. Drawing on
the social exchange theory and using Federal Employees Viewpoints Survey (FEVS) from 2011-
2015, this study expected that work-life balance policies would have significant positive
relationship with perceived organizational performance in the U.S. federal agencies, and that
leadership support will have a significant positive moderating effect. The study also expected
that the relationship between work-life policies and perceived organizational performance will be
more positive in feminine organizations compared to masculine organizations. However, the
findings do not support all these expectations. With the exception of employee’s assistance
programs, work-life balance policies do not seem to have a significant positive impact on
perceived organizational performance. In terms of moderation effect, the results indicate that
leadership support only has a significant positive moderation effect on the relationship between
perceived organizational performance and childcare programs, alternative work schedules, and
wellness programs in all U.S. federal agencies. Leadership support also has a significant positive
moderation impact on the relationship between perceived organizational performance with
alternative programs, and wellness programs in feminine organizations. This study asserts the
need to re-evaluate the implementation and the practices of work-life balance policies in the
federal agencies. This study also encourages public administration scholars to conduct more
systematic research on work-life balance policies to provide more concrete evidence on its
importance in the public sector context, and to provide recommendations on how to improve the
impact of such policies
Forecasting Stock Price Movements and Stock Trading Automation Using Deep Learning and Reinforcement Learning
No full textArtificial intelligence algorithms and big data analysis approaches are becoming more signif-
icant in a variety of application fields, including stock market trading and automation. Few
research, on the other hand, have concentrated on predicting the future directional change in
stock prices, particularly when utilizing strong machine learning methods like deep recurrent
neural networks (DRNNs) to conduct the analysis. For algorithmic trading and investment
management, developing a forecasting system that accurately anticipates future changes in
a stock price is critical. To address that, we propose a hybrid deep learning model with
self attention mechanism, dense layers, and a stacked bidirectional long-short term memory
neural network.
Many scholars have used technical analysis for financial forecasting with great success. When
computing the technical indicators, a time horizon parameter needs to be specified as an
input window size. In this work, the input size is set same as the prediction horizon or time
step. This is due to the fact that the stock price’s behavior over a prediction horizon may,
to some extent, mirror its previous behavior over the same time period.
For extracting temporal features from stock sequential data, a stacked bidirectional long-
short term memory neural network is proposed. The self-attention mechanism which directs
the neural network to place more weight on important temporal information is also proposed.
Following the model evaluation methods, experiments demonstrated that the proposed model’s
trading strategy is better than the buy-and hold trading strategy and the model also out-
performs other state-of-the art learning algorithms based on the evaluation metrics.
Majority of trades are now entirely automated, and algorithmic stock trading has become a
standard in today’s financial market. In many difficult games, like Go and Chess, Reinforce-
ment Learning (RL) agents have shown to be a formidable opponent. The historical prices
and movements of the stock market may be seen as a complicated, chaotic and imperfect
environment in which we aim to optimize return while minimizing risk. Three deep rein-
forcement learning agents are trained and an ensemble trading strategy is obtained using
policy gradient based algorithms: Deep Deterministic Policy Gradient (DDPG), Proximal
Policy Optimization (PPO) and Soft Actor Critic (SAC).
To improve the robustness and accuracy for representing stock market conditions, the Long
Short Term Memory (LSTM) is proposed to extract important and informative features from
raw financial data and technical indicators. The ensemble approach takes the best elements of
the three techniques and combines them, allowing it to adapt to changing market conditions
with ease. We use a load-on-demand strategy for processing extremely big data to prevent
significant memory usage in training networks with continuous action space. The algorithms
are tested on the 30 Dow Jones equities. The trading agent’s performance is assessed and
compared to the Dow Jones Industrial Average index and the classic min-variance portfolio
allocation approach. In terms of risk-adjusted return as evaluated by the Sortino ratio, the
proposed deep ensemble method outperforms the two baselines and a state-of-the-art deep
reinforcement algorithm
Pricing Residential Electricity in the Presence of Distributed Solar Energy Generation, Limited Consumer Response and Microgrids
No full textThis dissertation studies pricing of electricity in residential electricity markets. Among
all the sectors (residential, commercial, industrial, and transportation) of the USA power
market, the residential sector is the largest, with total revenue reaching more than 187 billion
dollars in 2019. However, at the same time, this market is challenging to manage for the
utilities primarily due to the growing supply and demand variability. Generation (often
via intermittent renewable resources) of electricity by consumers and supply of the excess
generation to the (main) grid induces supply variability. Extreme weather events (climate
change) and limited attention of consumers to these lead to demand variability. In response
to these challenges, local electricity generation and its local consumption are emerging as
potential strategies, which are implemented in the form of microgrids. A microgrid is built
in addition to the existing grid and it executes electricity purchase/sell transactions with the
grid.
Pricing of electricity in the presence of these fundamental demand and supply changes is
an important issue. This dissertation consists of three main chapters to address this issue.
In Chapter 2, we provide a revenue maximization formulation for a regulated utility and
reveal the interaction between rising optimal prices and growing solar power adoption. This
interaction can significantly reduce the number of customers of a utility and is referred to as
the utility death spiral in the power industry. We propose electricity pricing mechanisms to
slow down and stop this spiral. In Chapter 3, we propose a novel framework that formalizes
a household’s electricity consumption through setting appliances (e.g., an air conditioner) to
various levels. Focusing on the consumption decision making process, we provide a theoretical
foundation for analyzing a consumer’s limited capability in responding to changes in her
ambient environment.
In Chapter 4, we consider a microgrid’s capacity, its excess demand/supply and the resulting
transactions with the grid. The microgrid’s profit is formulated by taking the statistical
dependence between the market demand and price into account and also by ignoring this
dependence. We reveal the differences between these two profits and the optimal microgrid
capacities they lead to. Our results shed some light on why microgrid investments made
without incorporating the dependence can be disappointing
Virtual Tools Based on Medical Imaging and Mechanical Modeling to Optimize the Soft Tissue Envelope in the Transfemoral Residual Limb
No full textAn estimated 185,000 Americans sustain limb amputations every year due to trauma, cancer or
vascular diseases such as diabetes. Due to the increased number of people living with obesity and
diabetes, the incidences of major upper and lower limb amputations are projected to continue to
grow. To improve the quality of life of these individuals such as their abilities for self-care,
community integration, employment, and participation in activities of leisure, engineers seek to
advance technologies that either improve their abilities or better inform guidelines for their clinical
care. These advances in technologies include personalized designs of passive prostheses, intuitive
control interfaces of mechanically-active (i.e., robotic) prostheses, and customized and adaptable
prosthetic socket solutions, to name a few. With the drive for improved prosthetic technology, it
is vital to develop a deeper understanding of how the interface between the assistive device (i.e.,
the technology) and the human enables or curtails locomotor ability. In these studies, we examine
how the mechanical interface between prosthetic legs and the residual (i.e., amputated) limb can
be quantitatively assessed and optimized—focusing on techniques that target the limb and its
underlying structure independent from the effects of distal components such as prosthetic sockets,
knees and feet. Specifically, we focus on assessing and optimizing the soft tissue envelope within
the residual limb of individuals with transfemoral amputation using medical imaging, mechanical
modeling, and random sampling methods to inform clinical interventions that include surgical
recontouring of the limb and its underlying soft tissue. These tools could arm orthopedic and plastic
surgeons with quantitative and objective measures when considering the optimal limb “design” for
a given individual. This dissertation research contains three Specific Aims that observe (Aim 1),
gather data (Aim 2), and inform (Aim 3) how the form of the residual limb and its underlying
distribution of soft tissue influence its mechanical properties, and thus its abilities to don a
prosthesis
An Approach to Rapidly Assess Sepsis Using Machine Learning Approach
No full textSepsis is a life-threatening condition and understanding the disease pathophysiology using host
immune response biomarkers is critical for patient stratification. Lack of accurate sepsis
endotyping impedes clinicians to make timely decisions alongside insufficiencies in appropriate
sepsis management. The objective of this work is to demonstrate the potential feasibility of a
data-driven validation model for supporting clinical decision to predict sepsis host-immune
response. Herein, we used machine learning approach to determine the predictive potential of
identifying sepsis host immune response for patient stratification by combining multiple
biomarker measurement from a single plasma sample. Results were obtained using the following
cytokines and chemokines IL-6, IL-8, IL-10, IP-10, TRAIL, PCT and CRP where the test dataset
was 70%. Supervised machine learning algorithm naïve Bayes and decision tree algorithm
showed promising accuracies of 96.64% and 94.64% respectively. Using unsupervised clustering
algorithms, we are able to achieve silhouette score of positive 0.5. These promising findings
indicate the proposed AI approach could be a valuable testing resource for promoting clinical
decision making
A Fish Out of Water?: How Literary Theory Can Benefit Legal Interpretation
No full textThe correct theory of legal interpretation courts should apply to the cases and controversies
before them has been the subject of long and often heated debate. Such argument has not
progressed to any conclusion in over 200 years. This dissertation proposes to apply literary
critical theory to the analysis of various legal interpretive approaches in order to: (1) understand
why the debate is futile; and (2) propose the method of legal interpretation that presents the best
opportunity for the courts to make transparent and comprehensible decisions based on a
combination of text, history, technical, and social development, as well as what is best for the
public and for the cultural or ethnic group whose rights are at stake.
In order to accomplish this task, this work examines sample U.S. Supreme Court cases of social,
legal, or political significance over the last six decades. In doing so, this dissertation analyzes the
interpretive legal theory the author of the majority opinion claims to apply to the decision in
order to determine whether the claimed theory and the opinion coincide. This investigation
suggests a frequent disconnect between the theory claimed and the holding of the opinion.
Literary theory is then applied in an effort to explain this disconnect. On the basis of this audit,
this dissertation proposes the legal interpretive method that appears most consistent, transparent,
and ethical, and argues for the routine application of that methodology by the courts.
The research reviewed suggests that literary critical theory proves extremely helpful in
explaining the apparent disconnect between theory and result in Supreme Court opinions. The
work of scholars such as Stanley Fish, Steven Knapp, and Walter Benn Michaels suggests that
each of us thinks through a filter of our personal experience, education, religious beliefs, ethnic
and cultural group/status, political views, and expectations. Because we interpret events through
this filter, it is impossible for any justice or judge to impartially apply a theory that exists
independently of their system of beliefs. This conclusion is based largely on the literary reader response theory – developed by scholars like Hans Robert Jauss, Stanley Fish, and others –
which posits that each reader finds meaning in a work based not only on the text but also on the
basis of his history, experiences, expectations, knowledge, and beliefs.
Having examined a wide variety of opinions based on different interpretive approaches, this
work advances the legal interpretive method that seems not only to work most consistently in its
analysis, but also results in decisions that are most often transparent, ethical, just, and in the
current best interests of the public and its various ethnic and social minorities. That approach –
which I refer to as “ethicism,” but is also known as “living constitutionalism” – allows the court
the flexibility to consider not only the words and historical meaning of a text, but also changes
over time in society, its values, and technology that can impact the outcome. Although this
methodology is criticized as offering the Supreme Court excess authority and the ability to usurp
the powers of the legislative and executive branches, this work argues that such authority is
already vested in the Court by the Constitution and precedent.
This dissertation concludes that the endless debate over legal interpretive theories is largely futile
because justices and judges are largely unable to apply them neutrally and without prejudice. It
argues that a better approach is recommended by literary theory: ethicism, as informed by reader
response theory. Furthermore, the increasing politicization of the Court has eroded public trust in
the institution as fair and impartial. This fact threatens the Court with changes by the legislature
or executive in the form, makeup, or selection process of the Court. Ethical decision making
could help restore that trust.
There could be important potential social consequences were the Court to adopt of a consistent
ethicist method of deciding cases. This work argues that oppressed minorities and opposition
groups could benefit from a fairer, more objective approach by the Court, as demonstrated by the
Black Lives Matter and MeToo movements and critical cases such as the upcoming review of
Roe v. Wade. Society is harmed by the inconsistency and obvious prejudices of the Court, and
adoption of ethicist methodology could ameliorate that harm
The Voices of the Terrorized
No full textThis MFA thesis is about spreading the awareness of domestic violence that has risen since the
beginning of the covid-19 pandemic. Photo montage videos will exhibit a variety of notecards
that will shed light on a glimpse of survivors' or their confidants’ thoughts after or during a
critical predicament. By including graphic design images, I will show statistics on the drastic rise
of domestic violence during and after the covid-19 pandemic. Through this creative project, I
iterate on my initial participatory installation (2020). My installation creates a connection with
survivors, their confidants, and viewers
Towards Faster Software Revision Testing
No full textSoftware systems have been increasingly prevalent in all facets of our lives over the last few
decades and play a critical role in modern living. They have a significant impact on the
quality of our lives and provide tremendous convenience. However, software faults (also
known as bugs) are unavoidable throughout the development of software systems that can
have a substantial negative impact on the commercial company and result in significant losses.
Numerous researchers have been working on this problem to test software systems during
development and fix bugs after the software systems are established. However, due to the
complexity of these systems, these approaches can be very time consuming. For example,
mutation testing is an important component of software testing which can be very powerful
to evaluate the quality of the test suite, but it can be extremely time consuming due to a
large number of mutant execution. Also, Automated Program Repair (APR) techniques can
reduce software debugging human efforts by advising plausible patches for buggy programs.
However, the APR techniques need to repeatedly execute all the test suites to identify the
plausible patches for the bugs under fixing. This process could be extremely costly. Therefore,
it is essential to explore some approaches to speed up the processes of software testing and
debugging.
In this dissertation, we aim to speed up software testing and debugging via faster software
revision testing. The idea is to decrease the testing time between different revisions to speed
up software testing and debugging. We explored two scenarios in software testing during the
evolution of software systems: mutation testing and behavioral backward incompatibilities
(BBIs) detection. We applied regression test selection (RTS) techniques to speed up mutation
testing for the first study. Our study showed that both file-level static and dynamic RTS could
achieve efficient and precise mutation testing, providing practical guidelines for developers.
We called the second BBIs detection technique DeBBI which can reduce the end-to-end testing
time for detecting the first and average unique BBIs by 99.1% and 70.8% for JDK compared
to naive cross-project BBIs detection. Additionally, we detected 97 BBI bugs including 19
that were previously confirmed as unknown bugs. Lastly, we explored the application in patch
validation of APR technique to speed up software debugging. We treated every single patch
as a revision to develop a unified on-the-fly patch validation framework, named UniAPR. Our
study demonstrated that on-the-fly patch validation could often speed up state-of-the-art
source-code-level APR by over an order of magnitude, enabling all existing APR techniques
to explore a more extensive search space to fix more bugs in the near future
Molecular Beam Epitaxy of La2-xSrxCuO4 Films and Heterostructures
No full textSince 1986, the study of high-temperature superconductivity (HTS) in cuprates has revealed a
massive amount of discoveries, such as pseudogap, charge density wave, d-wave
superconductivity, etc. These novel states of matter trigger even more unknowns in fundamental
science and inspire enormous emergent applications. This dissertation presents our research on the
archetypical La2-xSrxCuO4 (LSCO) thin films and heterostructures. Specifically, the research has
been driven by several fundamental questions. For example, can we create c-axis Josephson
junctions for scientific research and superconductor-based quantum computation? What controls
the fundamental behaviors of interface superconductivity? To answer those questions, high-quality
crystals are required. Here we utilized and improved the oxide atomic-layer-by-layer molecular
beam epitaxy (ALL-MBE) technique to grow atomically smooth cuprate films and heterostructures
to answer the proposed research questions.
The main results are presented as follows. First, we improved the ALL-MBE growth in several
ways to enhance the film quality significantly. Specifically, we studied the thermal annealing of
oxide substrates and developed treatment methods for LaSrAlO4(LSAO) and SrTiO3(STO)
substrates. Ramp-up rate and annealing temperature are found to be the most critical parameters.
We then studied the synthesis of LSCO thin films via the ALL-MBE system. A detailed recipe for
the growth of LSCO thin films on LSAO substrates is presented. Unique reflection high energy
electron diffraction (RHEED) pattern features are observed in LSCO films. A strategy to monitor
the film growth and maintain the correct stoichiometry is developed based on the real-time
RHEED feedback. We also investigated the power and stability of ozone oxidation and compiled
empirical post-annealing procedures suitable for various doping levels. Substrates and LSCO films
were evaluated using atomic force microscopy (AFM) and RHEED. The results indicate that they
are atomically perfect with high crystallinity. Mutual inductance (MI) tests reveal that the LSCO
films are uniform over the whole sample area with a sharp superconducting transition. Second,
LSCO heterostructures and superlattices have been synthesized to study the HTS c-axis Josephson
junction and interfacial superconductivity. The method to probe the superconducting dead layer
number near the interface is introduced using a series of superlattices. At the LSCO-LSAO
interface, MI and transport measurements imply that the first two LSCO layers that are near the
LSAO exhibit a substantial suppression of superconductivity, resulting in a barrier that is five
layers thick in total. And an overdoped LSCO protective layer is found to be effective against
carrier depletion in superconducting layers. Within LSAO barriers, a thickness of 2 unit-cells of
LSCO interface superconductor is synthesized. The superconducting transition of the sample is
tunable with doping and demonstrates the highest transition temperature of 34 K
Wind Farm Flow and Power Capture: Optimal Design of LiDAR Experiments, Flow Physics, and Mid-fidelity Modeling
No full textNowadays there is an urgent need for wind farm flow models with increased accuracy and
low computational costs for the prediction of turbine performances and wakes. An improvement of current standards of wind farm simulations can be achieved only through a better
understanding and modeling of the complex physical mechanisms governing the wind farm
aerodynamics. Low computational requirements are necessary to enable large amount of
simulations needed for the optimal design, real-time monitoring and online control of wind
power plants. To this aim, a holistic research project has been conceived and implemented,
which is the focus of this Ph.D. thesis. The adopted research strategy includes three main
tasks: i) optimal design and execution of field experiments for monitoring wind-farm operations through scanning LiDAR, meteorological and SCADA data; ii) statistical analysis of
LiDAR field measurements for probing wake evolution, wake interactions, effects of atmospheric stability, and flow distortions due to topography; iii) development of a data-driven
RANS model for accurate and low-computational-cost simulations of wind farm operations.
This research project has enabled quantifying and modeling effects on wind farm operations
connected with the turbine aerodynamics and the atmospheric stability regime, and detecting the occurrence of topography wakes, which are flow regions with reduced wind speed
and enhanced turbulence intensity being detrimental for wind turbines installed on complex
terrains. The main deliverables of this project are the LiDAR Statistical Barnes Objective
Analysis (LiSBOA), a tool for the optimal collection and statistical characterization of LiDAR measurements, and the Pseudo-2D RANS (P2D-RANS) wind farm model, which has
been recently distributed among several industrial partners and with the intended uses of
simulating and monitoring the operations of several wind power plants