8413 research outputs found
Sort by
Face Verification with Veridical and Caricatured Images using Prominent Attributes
Caricatures, with their exaggerated features, offer a surprisingly efficient means for individuals to recognize each other compared to veridical (real) images. However, it is still a difficult task in machine learning to match veridical images to caricatures. This is due to the poor quality of caricature datasets, which often lack clear labels and contain low-quality images. Widely utilized veridical image datasets like CelebA also suffer from inadequate labeling. These label inconsistencies pose significant issues in accurate face verification tasks. Moreover, the effectiveness of neural networks has led to a shift in focus away from attribute-based representations, further complicating the matching process. In this thesis, we introduce a classification protocol for prominent facial feature recognition along with a verification protocol for matching celebrity veridical images to their caricatures. We utilize CarVer, a recently curated dataset comprising both veridical and caricature images accompanied by detailed prominent feature labels
Unraveling the Function and Regulation of Circular RNAs in Aging and Age-Related Diseases
Aging is a time-related deterioration of the physiological functions necessary for survival and fertility and is associated with changes in dynamic biological, physiological, environmental, psychological, behavioral, and social processes. Among many factors that might play a role in aging and age-related diseases, circRNAs have been linked to aging and Alzheimer's Disease (AD). circRNAs are non-coding RNAs produced via a back- splicing event and expressed in many organisms, but their function in aging and age- related diseases remains unclear. My work presented in this dissertation is focused on understanding the role of circular RNAs in aging and AD development. I used Caenorhabditis elegans as a model organism to investigate the role of circRNAs in aging and AD. Previous work showed that circRNAs accumulate during aging in several organisms, including C. elegans . We hypothesize that circRNA accumulation during aging is detrimental and might contribute to the progression of age-related diseases such as AD. In Chapter 2 , we used a CRISPR/Cas9 strategy to genetically remove circRNAs (referred to as circ- crh-1 ) derived from the host crh-1 /CREB gene by targeting intronic sequences that are required for circularization. We found that mutations in circ- crh-1 increased the mean lifespan of C. elegans . In Chapter 3 , we showed that loss of this age- accumulated circ- crh-1 ameliorated the amyloid b-induced toxicity in a well-established C. elegans transgenic model for AD. We further show that the collagen gene, col-49 , functions in promoting amyloid b-induced paralysis. Lastly, in Chapter 4 , we investigated the function of the RNA-binding protein NOVA-1 in C. elegans . Its homolog in mice, NOVA-2, is a known regulator of circRNA biogenesis in neural tissues of mice. In C. elegans , we found that mutations in nova-1 extend the mean lifespan and show increased heat resistance. Moreover, we identified both new and existing circRNAs that are differentially expressed in nova-1 mutants, including circ- crh-1 . Together, these results suggest that NOVA-1 might regulate lifespan by controlling the abundance of a sub-group of circRNAs. However, further experiments are required to understand better the role of NOVA-1 in circRNA regulation and the aging process
Modeling Quantum Tunneling in Nonadiabatic Statistical Theory
The central topic of this work is predicting the kinetics of spin-dependent processes. In the first part of this dissertation, we provide an overview of the nonadiabatic statistical theory (NAST) and our NAST program package for predicting the rate constants of spin-dependent processes. We also compare the capabilities of NAST with those of the nonadiabatic molecular dynamics (NAMD), a complementary approach for studying spin-dependent processes. This comparison is supplemented with several examples to highlight the advantages and limitations of both approaches. In the second part, we focus on the application of NAST to study the kinetics of spin-dependent processes in the quantum tunneling regime. Comparison of the semiclassical Landau-Zener and quantum mechanical weak coupling rate constants allows us to estimate the role of quantum tunneling in reaction kinetics at low temperatures. We use the model of linear crossing potentials to investigate how molecular properties, such as energy gradients, spin-orbit coupling, and mass affect the kinetics of a spin-dependent reaction in the tunneling regime. We extend this work to study the kinetics of spin-forbidden isomerization reactions in three complexes of Ni(II) and two spin-crossover reactions in nitrenes at cryogenic temperatures. We also use these molecules to test the performance of a new Landau-Zener-Eckart method for describing quantum tunneling in spin-dependent reactions. The third part of this dissertation is motivated by the discussion of the ground state multiplicity of iron(II)-porphyrin, the prosthetic group of hemeproteins. The ground state of iron(II)-porphyrin remains under debate due to the close-lying electronic states of different spin multiplicities arising from the partially filled 3d shell of iron. We propose the existence of a nonadiabatic equilibrium between the lowest energy triplet and quintet spin states. The last part of this dissertation is motivated by the remarkable ability of vulpinic acid, a natural sunscreen pigment found in lichens, to protect the organism against harmful UV radiation. The photostability of vulpinic acid drastically increases when it is embedded in a polysaccharide matrix. We propose that this enhanced photostability originates from the formation of a H-bonding network between the pigment and the matrix. We perform computational studies on the ground and excited electronic states of vulpinic acid that are proposed to participate in the energy dissipation mechanism behind the photoprotective activity of vulpinic acid
Desert Conservation Priorities Mapping Tool
Development of public land use within the Desert Renewable Energy Conservation Plan (DRECP) boundary is taking place through collaboration of the California Energy Commission (CEC), the U.S. Fish and Wildlife Service (USFWS), the California Department of Fish and Wildlife (CDFW) and the U.S. Bureau of Land Management (USBLM) in order to 1) identify areas of land with high-quality renewable energy potential and 2) provide for the long-term conservation and management of endangered and threatened species as well as areas of biological, physical, cultural and social importance. The DRECP covers 22,585,000 acres of southern California, encompassing the Mojave and Colorado Deserts. Home to threatened, endangered and protected (TEP) species, endemic lineages and evolutionary hotspots, the desert biota face widespread threats (e.g. human development, subsidized predation, linear corridors, off-highway vehicles, military training, invasive plants, climate change, and renewable energy). Privately owned land existing within the DRECP provides potential opportunities for private environmental organizations to expand on federal and state landscape-scale conservation efforts. I will be showcasing a spatially explicit decision support system (DSS) that was built to help a private environmental organization identify parcels to target for land acquisition, conservation, and restoration. Differing from others in its versatility and its weighting scheme, this DSS works under differing conservation scenarios (e.g. conservation vs restoration) and includes nonlinear weighting of criterion (attribute) values, providing more flexibility to the decision maker when assigning weights to evaluation criteria
Synthesis, Fabrication and Functionalization of Two-Dimensional Polymers
The synthesis of two-dimensional polymers (2DPs) is an emerging field with expected applications in nano-electronics, separation membranes, and memory storage devices. Unlike linear or branched polymers, 2DPs are defined as topologically planar sheets having a thickness consistent to their monomer species. This dissertation revolves around the synthesis, fabrication, and surface functionalization of a 2DP coined poly(carboxy fantrip).Chapter one focuses on the definition, synthetic approaches, and applications of 2DPs. Both natural and synthetic two dimensional (2D) materials are reviewed, including different polymerization approaches, transfer processes and, surface modifications. Chapter two reports the optimization of a 2D monomer, carboxy fantrip, and derivatized versions on a preparative scale. This chapter addresses inconsistencies from previous procedures increasing monomer yields from a few milligrams to hundreds. Synthetic strategies, both working and unsuccessful are reported, including experimental protocol and characterization of intermediates. Chapter three discusses new photo-polymerization strategies at the air/water interface, polymer deposition onto solid substrates, and covalent surface modifications of the resulting film. The results of these experiments are monitored and analyzed by Brewster angle microscopy (BAM), differential interference contrast (DIC) microscopy, high speed videography, and contact angle measurements
Whole Grains and Phosphorus in Chronic Kidney Disease
Chronic kidney disease (CKD) is one of the leading causes of death in the United States and impacts around 15% of the population. The development and progression of CKD introduces the risk of developing CKD-mineral and bone disorder (CKD-MBD), which has been attributed to excess dietary phosphorus intake. As a result, this has led to people with CKD being advised to restrict whole grains and consume refined grains, which are lower in phosphorus. However, whole grains are generally recognized as being healthier, and there is no direct evidence linking whole grains to kidney decline or CKD-MBD. This dissertation will address three research questions: 1) to what extent does restricting whole grains lower phosphorus exposure in people with CKD; 2) what are the effects of whole grain consumption on CKD-MBD and mortality; and 3) what are the perspectives of people with CKD regarding whole grains in their diet. Study 1 (Chapter 2). The contribution of whole grains to phosphorus exposure and CKD-MBD was explored and analyzed through a narrative review. After evaluating known determinants of phosphorus exposure from grain products, three model diets were created based on the reported grain intake of the general U.S. population, and adjusting for the likely contribution of phosphate additives, as well as differences in phosphorus bioavailability: standard grain diet, 100% refined grain diet, and mixed (50/50 whole and refined grain) diet. This analysis found that the additional phosphorus exposure from a mixed grain pattern (231 mg/day) compared to the 100% refined grain (127 mg/day) was unlikely to be enough to raise serum phosphorus levels in most people with CKD and could be meaningfully mitigated by avoiding phosphate additives in processed grain products. Study 2 (Chapter 3). The association of whole grains with CKD-MBD and mortality in people with CKD was analyzed in the Chronic Renal Insufficiency Cohort (CRIC) study. The cross-sectional analysis found no strong association between whole grains and CKD-MBD biomarkers. Although, the prospective analysis found that whole grains in relation to refined grains and overall energy intake were associated with a lower risk of cardiovascular disease (CVD) and mortality. Study 3 (Chapter 4). Reported intakes of whole grains and participants perspectives regarding grain consumption were assessed in a cross-sectional study of patients undergoing hemodialysis treatment in Northern Nevada. The majority of participants did not report dietitians as the most reliable source of information and although most participants stated they had some form of confidence about grain recommendations, their opinions differed, with most stating people with CKD should consume less grains than the general population. However, some participants stated that people with CKD should consume whole grains. Conclusions. The results of this dissertation do not support the current practice of restricting whole grains and support a link between how the lack of agreement about grain recommendations may lead to confusion among people with CKD
Essays in Economics: Analyzing Homeowners’ Willingness-to-Pay for Wildfire Risk Mitigation, and the Overlapping Generations (OG) Model of Interest Rate Dynamics.
This dissertation examines the estimation of homeowners’ willingness-to-pay for wildfire risk reduction and the determination of interest rates. The first chapter employs a contingent valuation approach to estimate homeowners’ willingness-to-pay (WTP) for wildfire risk reduction in 35 Wildland Urban Interface (WUI) communities in Nevada. Survey respondents were presented with two potential risk reduction programs: A private program focusing on individual home and surrounding vegetation modification, and a public program targeting community-wide risk through fuel management treatments. In this study, we employed a contingent valuation approach to estimate homeowners’ willingness-to-pay (WTP) for wildfire risk reduction in 35 WUI communities in Nevada. We presented respondents with two potential risk reduction programs: a private program focusing on individual home and surrounding vegetation modification, and a public program targeting community-wide risk through fuel management treatments. We found significant WTP for private program risk reduction but not for public program risk reduction. Our work also identifies a number of factors that my bias WTP estimates if not considered. These include respondent’s previous expenditure on risk mitigation, the possible loss level from fire, concern about the non-financial costs of risk mitigation, and recognition of the communal (i.e. positive externality) benefits of fire risk reduction. The second chapter presents the Weil (1987) overlapping generations (OG) model, which contains an exogenous probability that the economy’s bubble may burst but extended to allow capital accumulation as in Banerjee (2021). Like Weil (1987), we find that the rate of return on the bubble asset must generally be greater than the rate of return on the capital backed asset. Like Banerjee (2021), but in contrast to Weil (1987), a gap between the interest rate paid on the capital backed asset and capital rental rate must occur. Thus, we provide enhanced knowledge of how the rates of return earned on assets relate to the productivity of capital and the capital rental rate.The third paper extends the OG model of Weil (1987) by adding capital accumulation and stock market clearing as introduced in Banerjee and Pingle (2023). The addition of the stock market clearing eliminates the indeterminacy and inefficiency from Weil's model, resulting in a unique, Pareto efficient equilibrium. Because the bubble may burst in our model, as in the Weil model, we like Weil find that the rate of return on bubbly assets must exceed that on capital backed asset to offset the risk of bubble bursts. In contrast to Weil, we find that the path for the bubble is not influenced by a change in the perception that the bubble will burst. The bubble will form if it can form, and it is the rate of return on the bubble asset that will adjust in response to an increased perception that the bubble will burst
Conflict-Based Pedestrian Safety Scoring With All-Traffic Trajectory Data
The increasing urgency for pedestrian safety in urban areas underscores the necessity for advanced methodologies in traffic management, aligning with the Vision Zero initiative's goal to eradicate traffic-related fatalities and severe injuries. This dissertation introduces a novel model that utilizes all-traffic trajectory data obtained from LiDAR sensors to enhance pedestrian safety assessments at urban intersections. By leveraging a comprehensive set of variables�"including near-miss incidents, pedestrian volumes, vehicular speeds, trajectory data, temporal patterns, lighting conditions, and crosswalk visibility�"this model provides a holistic scoring methodology for pedestrian risk evaluation. Central to the model's efficacy is the application of the Post-Encroachment Time (PET), a critical temporal metric that measures the time interval between potentially conflicting trajectories, thus allowing for the anticipation and mitigation of collision risks. This innovative approach not only proposes a shift from reactive to proactive safety measures but also aims to establish a new approach to utilizing cutting-edge pedestrian data for urban safety enhancements. The findings from this study are poised to offer significant insights to urban planners, policymakers, and transportation engineers, aiming to cultivate safer pedestrian environments in growing urban landscapes. In the case studies of applying the proposed scoring system, twelve intersections were selected for data collection and pedestrian safety scoring with the proposed method. Many of the studied sites with high-risk scores also had larger numbers of crashes, showing a possible positive correlation could be discovered as more locations are completed. This research initiates the foundation of a conflict-based pedestrian safety scoring system, which can be extended to accommodate more impacting factors, such as time of day, existing pedestrian safety treatments, pedestrian behavior patterns, and land uses. Analysis of correlation between pedestrian risks and those impacting factors with the innovative trajectory data can also initiate new research topics based on what is introduced in this dissertation
Supporting LGBTGEQIAP+ Community College Students: Associations of Campus Resource Utilization and Campus Climate with Psychological Well-Being
LGBTGEQIAP+ people experience mental health distress at higher rates than cisgender heterosexual people, which can be connected to added layers of stress due to patterns of negative treatment on the basis of their identities (Meyer, 2003). LGBTGEQIAP+ community college students are at especially high risk for experiencing challenges, psychologically, socially, and academically compared to cisgender heterosexual students. This dissertation explores LGBTGEQIAP+ campus climate and on-campus resource utilization as potential contributors to psychological well-being, instead of distress, which encompasses different dimensions of positive mental functioning (Ryff, 1989a). Survey data from 103 LGBTGEQIAP+ community college students was used to analyze linear regression models. Results showed some evidence of connections between resource utilization and psychological well-being (RQ1) and significant associations between aspects of LGBTGEQIAP+ campus climate and psychological well-being (RQ2). Implications include the importance of community colleges addressing LGBTGEQIAP+ issues on campus directly, attending to concerns about safety on campus, and increasing the availability of LGBTGEQIAP+ specific supports on campus, with recommendations of how community college counselors can lead such efforts. Limitations and future directions are discussed
On Normal Maps and Noether's Isomorphism Theorems for ∞-Groups
In this thesis we further develop the notion of homotopy normal maps of loop spaces introduced by Prasma. This differs from the classical notion of normality in group theory in that the property of being a normal subgroup is replaced with a normality datum. In particular, we investigate the relationship between these ideas and Noether's three isomorphism theorems in classical group theory. We show that an analog of Noether's first isomorphism always holds; the second does not hold in general; and we provide sufficient conditions for which the third isomorphism theorem holds in the context of ∞-groups. In the process of proving these claims we develop a factorization system on the ∞-category of ∞-groups and relate it to the classical epimorphism/monomorphism factorization of ordinary groups