33760 research outputs found
Sort by
Categorification of the Leray spectral sequence
No full textThe first part of this thesis concerns the categorification of differentials in the Leray spectral sequence. Given a continuous map of topological spaces and a sheaf of abelian groups , the Leray spectral sequence takes the form
We provide a categorification of certain differentials in this spectral sequence, specifically those of the form
by constructing a gerbe that encodes the first obstruction to a lifting problem.
The differential is first described at the level of Čech cocycles. Then, for a class , we construct the sheaf of local lifts, which takes values in -groupoids. Taking the 1-truncation of this object yields an -banded gerbe whose cohomology class represents the image of under . Finally, we show that this construction is compatible with the additive structure on cohomology.
Second part concerns it with the existence of non-cellular objects in the motivic stable homotopy category
THREE REEDS & A HORN: EXPLORING TRIOS FOR CLARINET, HORN, AND BASSOON
No full textOriginally founded from a lighthearted collaboration between woodwind quintet members, trios for clarinet, horn, and bassoon, known as “Three Reeds & a Horn” ensemble has captured my interest as a performer and collaborator. Comprised of 3/5ths of a woodwind quintet, this trio is not a standard chamber ensemble. However, the repertoire from the Western European canon dating from the late 1700s – to early 1800s, would suggest that this ensemble has significant value and was a distinctive chamber ensemble. Through this dissertation, I seek to provide greater visibility for the Clarinet, Horn, & Bassoon trio ensemble, to highlight its legitimacy and worthiness of being studied, programmed, and performed. The available repertoire from the late baroque to early classical eras are excellent material for a lecture recital as the earliest surviving compositions for the trio originate from the late 1700’s, with increasing output through early 1800’s, which suggests these pieces were composed for pre-existing ensembles and likely popular. In addition to bringing forth historical works, this dissertation includes contemporary works, new commissions, and flex chamber music (where the instrumentation can be altered with interchangeable parts) which can easily be incorporated into a program with this ensemble. This flexibility creates a medium where 21st-century musicians can serve as advocates by commissioning underrepresented composers, such as women and composers of color, and incorporating themes to bring awareness and serve as advocates for causes such as gender inequality, climate change awareness, or civil rights advocacy. Finally, as music educators and professionals seek to increase access to chamber music for students and audiences, this ensemble provides new opportunities to program works spanning the gamut from the late baroque to the 21st century. The repertoire is rich and remarkably complex, the opportunities are boundless, and the future is bright for this tiny forgotten trio ensemble
Quantum Games, Graphs, and Gödel
No full textThis thesis explores the quantum extension of a fundamental theme in theoretical computer science: the interplay between graph theory, computational complexity, and multiprover interactive proof systems. Specifically, we examine the connections between quantum graph properties, computability theory, and entangled nonlocal games.Building on prior work that defined quantum variants of graph homomorphisms, isomorphism, and chromatic, independence, and clique numbers, we introduce quantum perfect matching properties for graphs. We characterize these properties by quantizing a classical relationship between perfect matchings and the independence number of line graphs.
We then investigate the complexity of determining whether a nonlocal game is perfectly winnable with an entangled strategy. While the landmark result ???* = ?? established the undecidability of this problem, we show that it is doubly undecidable, proving its completeness for the class Π2 in the second level of the arithmetical hierarchy. This result is achieved by further developing the iterated compression technique.
Additionally, we define a family of nonlocal games generalizing the CHSH game and analyze their rigidity properties using a novel noncommutative sum-of-squares framework. This approach allows us to prove rigidity for games with quantum values bounded away from 1.
Finally, we study synchronous strategies, which are used throughout the literature and the rest of this thesis. We look at synchronous strategies in less studied contexts of non-synchronous games and synchronous games which are not perfectly winnable
Celebrating a Milestone: 20 Years of Sharing and Transforming Access to Resources
No full textSince its founding in 2004, the Sharing and Transforming Access to Resources Section (STARS) has been the home for resource sharing professionals within the Reference & User Services Association, a division of the American Library Association. As STARS begins its third decade, two officers reflect on how the section has evolved and the ways in which it has supported the resource sharing community and contributed to its advancement over the past two decades. They also share where STARS is heading next and outline opportunities for resource sharing and other access services professionals to engage and collaborate.https://doi.org/10.1080/15367967.2025.256603
ASSESSING OCEAN SALINITY AS NATURE’S RAIN GAUGE
No full textChanges in the hydrological cycle can have profound impacts on society, from more frequent and severe droughts to extreme flooding. However, monitoring changes in the hydrological cycle is difficult from current observing platforms (namely satellites), and even more difficult to identify secular changes in the often-noisy data. Near-surface ocean salinity patterns mirror the distribution of evaporation and precipitation over the ocean: regions dominated by evaporation are saltier, while regions dominated by precipitation are fresher. Recently, ocean salinity has gained attention as a proxy for tracking global hydrological changes. Altering salinity in the ocean through changes in the hydrological cycle can impact global ocean circulation, which could have major downstream impacts on global climate. Thus, this research focuses on the relationship between changing ocean salinity and the global hydrological cycle. The first chapter provides details on what ocean salinity is, why it is important, how it is measured, and what research gaps are addressed in this dissertation. The second chapter focuses on validating satellite-based surface salinity measurements with in situ observations to ensure that the global signals identified by satellites are reliable. The third chapter examines how surface salinity relates to evaporation and precipitation in the dynamic and climatically important North Atlantic. The fourth chapter leverages a vigorous methodology designed to minimize sampling biases and properly preserve and propagate uncertainties to estimate robust salinity pattern amplifications (salty gets saltier, fresh gets fresher) over short ( 60 years) time periods. We find salinity patterns have amplified at a rate of 4.89% per 50 years over the 1957/61 – 2019/23 pentadal record. Furthermore, we identify a 30–40-year period for when secular changes are identifiable (e.g., salty areas become saltier and fresh areas become fresher), and we detect and quantify an acceleration in the salinity pattern amplifications which may be indicative of an acceleration in the amplification of the hydrological cycle.
Finally, the fifth chapter addresses the future work we believe is critical to further our understanding of Earth’s climate through the lens of ocean salinity. One of the most important questions that has evolved from this dissertation is how will salinity impact the Atlantic Meridional Overturning Circulation (AMOC) in a warming climate. We identified changes in observed salinity that may help maintain/enhance the AMOC; however, it is unclear if increased meltwater and hydrologic amplifications will counter those changes. Thus, to address this question we must leverage both numerical simulations and observations. Since these changes are happening on multi-decadal time scales, it is critical that we continue to collect near-global ocean salinity measurements for the foreseeable future
Contextualizing Stigma: The Impact of School Cultural and Structural Contexts on Interpersonal Exclusion Following Criminal Justice Contact
No full textCriminal justice contact is often associated with weakened social ties, a process referred to as interpersonal exclusion (Jacobsen, 2020). Yet, most existing research assumes that the stigma of such contact is experienced similarly across settings, paying limited attention to contextual conditions that can foster or mitigate stigma. This study argues that context shapes stigma tied to criminal justice contact by influencing its perceptibility (i.e., the extent to which the contact is known to others) and dis-credibility (i.e., the extent to which the contact harms one’s social reputation). To capture both dimensions, the study introduces novel measures of cultural and structural context. Drawing on longitudinal data from a school-based study with rich adolescent peer network information and employing stochastic actor-oriented models, it examines how the cultural and structural contexts of grade cohorts in middle and high school influence adolescent friendship ties following police contact. It also explores whether the associations between police contact and friendship ties, as well as the moderating role of contextual factors, differ between middle and high school due to broader contextual distinctions. Findings indicate that police contact is associated with weakened adolescent friendship ties in middle school but not in high school. Cultural norms regarding the popularity of delinquent students moderate this association in middle school, whereas structural features of network closure moderate this association in both middle and high school. These findings underscore the contextual nature of stigma, highlight the role of cultural and structural environments in shaping interpersonal consequences of criminal justice contact, and offer implications for promoting the social inclusion of justice-involved youth
Hormone Care Is Healthcare: Addressing Access Gaps for Trans People in Maryland
No full textThe Maryland Trans Survey is a community-based research project conducted by Trans Maryland. The Queer/Trans Collective for Research on Equity and Wellness examining experiences of trans people in the State of Maryland in areas such as health and healthcare, employment and economic well-being, and legal and policy experiences. To date, it is the largest survey of trans people in the State, with 750 trans people representing all 23 counties in Maryland and Baltimore City.
Data were collected from June to December 2023 through in-person and online community outreach. The project was approved by Towson University’s Institutional Review Board (Protocol #1897) and used Transgender Research Informed Consent (TRICON) Disclosures to provide trans community members with additional transparency on the project, recognizing long histories of harmful practices in trans research from scientific institutions.
Transgender and nonbinary people face disproportionately high rates of depression and suicidality compared to their cisgender and heterosexual peers (Rastogi et al., 2025). Trans youth are 2 to 2.5 times more likely to report depressive symptoms and attempt suicide than cisgender LGBQ youth (Price‐Feeney et al., 2020). These elevations and mental health risks stem from societal rejection, stigma, and discrimination, not from their identities. One study also found a direct correlation between anti-trans laws and a 72% increase in suicide attempts among trans youth (Lee et al., 2024). Mental health struggles among trans adults mirror the experiences of youth. According to the 2022 U.S. Transgender Survey, 78% of respondents considered suicide and 40% attempted suicide at some point (Rastogi et al., 2025). These laws can include legislation that restricts access to gender-affirming care (GAC), including gender-affirming hormone treatment. While recent actions by the Trump administration have sought to restrict access to GAC by limiting the use of Federal funds, there have been no changes to Federal or Maryland state laws regarding GAC as of writing. Individuals in Maryland, including those covered by Medicaid, can still access gender-affirming care.
Gender-affirming care refers to a range of supportive services tailored to the needs of trans individuals. These services can include medical care, surgical procedures, mental health treatment, and non-medical forms of assistance. Gender-affirming hormone therapy is one aspect of gender-affirming care. It typically involves the administration of testosterone for individuals assigned female at birth and estrogen for individuals assigned male at birth. Gender-affirming healthcare, especially hormone therapy, has been shown to dramatically improve mental health outcomes, including lower rates of depression and suicidality (Green et al., 2022). To ensure trans Marylanders receive the care they need and deserve, it is essential to consider their experiences in relation to their access and use of gender-affirming hormones.
This brief presents project data related to Gender-Affirming Hormone Treatment for the purpose of helping advocates, policymakers, and community-serving entities to better understand and support the current needs of trans people in Maryland.University System of Maryland - Wilson H. Elkins Professorship (2021-2023); Washington University in St. Louis - Audre Lorde Distinguished Professorship (2023-present)https://transmaryland.org
Neural Networks as Databases - From Data to Model Compression
No full textThe past decade has witnessed an exponential increase in data and the rise of deep learning systems to handle it. These systems primarily analyze, learn and interpret the data by excelling in exploiting patterns.
This leads us to the question - can we use these patterns to efficiently store the data as well ?
Essentially this would mean moving from the paradigm of data compression towards model compression.
This thesis introduces and investigates a variety of methods in the realm of model compression,
with the eventual goal of replacing data compression itself.
To explore this, we start with the most information rich, yet sparse media form - videos. In the first part of this thesis, we introduce a framework for representing videos as
continuous functions using Implicit Neural Representations (INRs), which aim to represent any
given signal as a mapping between the spatial/temporal coordinate space to its values.
We propose an auto-regressive framework that exploits the redundancies of a video for
efficient compression and real time decoding - a first in this field. We then build
upon this framework by introducing a shared video prior that captures common patterns across video frames,
significantly improving the encoding times by 10-20, while improving the compression rates.
Despite these value additions, INR's aren't really representations of underlying data unless the resulting model weights also encode some semantics.
Based on this intuition, we introduce a hypernetwork-based INR system enables us to perform semantic tasks
like video retrieval and understanding along with compression.
Having established that the problem of data compression is actually a model compression problem, I will then present a case study on a widely used model compression method: pruning. We take a popular pruning method - the Lottery Ticket Hypothesis which offers extreme sparsity and study its effects on fundamental vision tasks like classification, detection and segmentation.
Finally, we take a look at the proposed future directions in which we explore enhanced network and algorithmic designs with random networks for greater compression and meta-learning for achieving faster video encoding
The Roles of Social Intimacy, Bias, and Group Norms in Children’s Intergroup Interactions
No full textChildren actively construct knowledge about their social world starting early in development. This knowledge includes an understanding about group norms, group dynamics, and group identity. Group identity includes an awareness of gender, race, and ethnicity. While group identity serves an important affiliative function, there are contexts in which biases emerge about other groups, often contributing to social exclusion and peer rejection. The present dissertation includes three papers which explore the impact of social intimacy, bias, and group norms on children’s attitudes and behaviors in intergroup interactions. Empirical Paper 1 explores whether social intimacy impacts children’s predictions and evaluations around interracial interactions. Children thought that high intimacy social contexts were more racially segregated than low intimacy contexts and more strongly rejected interracial exclusion in high intimacy social contexts, possibly due to high intimacy social contexts providing opportunities for quality contact and friendship formation. Empirical Paper 2 documents the role of includer direct bias and indirect bias justifications in children’s same-race inclusion evaluations and their moral reasoning. Children rejected forms of direct bias, but had more nuanced evaluations of indirect bias, and children’s moral reasoning was associated with rejecting complex forms of intergroup exclusion. Lastly, Empirical Paper 3 highlights the role of inclusive outgroup norms in children’s attitudes and behaviors towards gender outgroup members. Higher inclusive norm beliefs were associated with more intergender contact, more positive attitudes towards gender outgroup members, and inclusive norm beliefs appeared to be especially helpful for high-status children. Cumulatively, these three empirical paper provide novel evidence that children consider social intimacy, bias, and group norms, when making decisions about social inclusion and exclusion in intergroup interactions. The broader impact of this line of research lies with the potential to develop intervention programs designed to ameliorate early forms of prejudice and bias in childhood, which best prepares children for positive social relationships and healthy social development
Robust and Flexible Methods for Small Area Estimation
No full textSample surveys are widely used to provide estimates for both the overall population and various subpopulations, known as domains, which can be defined by geographic or socio-demographic characteristics. Direct estimators rely solely on domain-specific sample data and are typically design-based, incorporating survey weights and relying on the probability distribution induced by the sampling design for making inferences. Although the total sample size in a survey is typically large, the sample size for specific domains may be small or even zero. When a domain-specific sample is too small to produce direct estimates with adequate precision, the domain is classified as a small domain or small area.
The increasing demand for small area statistics has driven the development of small area estimation (SAE) techniques, which produce reliable estimates for domains with limited or no sample data. This dissertation focuses on enhancing the robustness and flexibility of model-based SAE methodologies by addressing three key challenges: model misspecification, flexible modeling, and uncertainty quantification.
The first study examines the effects of model misspecification on several commonly used small area estimators. The results show that when the underlying model is misspecified, the observed best prediction (OBP) method does not consistently outperform the Empirical Best Linear Unbiased Predictor (EBLUP) in terms of the design-based mean squared prediction error (MSPE), even though OBP being designed to improve design-based MSPE over EBLUP under model misspecification. Both analytical and numerical evidence are provided to show that OBP performs better when using aggregated auxiliary variables compared to using the individual ones. It offers practical insights for handling model misspecification in small area estimation.
The second study develops a framework for predicting complex small area characteristics, which are often nonlinear functions of the study variable for population units, using a nested error regression model with high-dimensional parameters. This study addresses multiple challenges simultaneously. First, it allows both regression coefficients and sampling variances to vary across areas, accommodating heterogeneity and enhancing modeling robustness. Second, we propose a novel algorithm for estimating area-specific model parameters, improving computational efficiency compared to existing algorithms. Third, we introduce a new approach for producing area-specific poverty estimates for out-of-sample areas, yielding less synthetic estimates than existing methods. Design-based simulation studies demonstrate that the proposed method outperforms existing approaches in terms of relative bias and relative root mean squared prediction error. Additionally, the method is applied to household survey data from the 2002 Albania Living Standards Measurement Survey to estimate poverty indicators for Albanian municipalities.
A measurement of any quantity of interest is complete only when accompanied by an evaluation of its uncertainty. The third study advances the theory of parametric bootstrap methods for constructing highly efficient empirical best linear (EBL) prediction intervals for small area means, incorporating both fixed and random effects. We analytically demonstrate that even when the normality assumption for random effects is relaxed, the proposed EBL prediction interval maintains a second-order coverage error, provided a pivot exists for a suitably standardized random effect when hyperparameters are known. In the absence of a pivot, we find that the order of coverage error of the parametric bootstrap EBL prediction interval is , and the first-order term is theoretically positive under certain conditions, indicating possible overcoverage of the EBL prediction interval. This characteristic may be advantageous for practitioners who do not account for other properties of prediction intervals. Furthermore, we propose a novel double bootstrap method, which can correct coverage issues in general. Monte Carlo simulations indicate that the proposed single bootstrap method performs well compared to alternative approaches.
Overall, this dissertation provides valuable insights into critical challenges in small area estimation, specifically in model misspecification, flexible modeling, and uncertainty quantification. Future research should explore semi-parametric and nonparametric methods to further enhance the robustness of inference for small areas