University of Illinois at Chicago
University of Illinois at Chicago: UIC INDIGO (INtellectual property in DIGital form available online in an Open environment)Not a member yet
21439 research outputs found
Sort by
The Determinants of Student Cross-Border Migration: An Applied Microeconomic Analysis
This dissertation examines two determinants of international student mobility, leveraging exogenous variation from quasi-experimental events to shed light on how students choose where to study. The first chapter estimates the impact of a shared bilateral education system on the location choice of international students. To analyze this, a two-way fixed effect model is derived from a conventional random utility maximization framework for optimal location choice. The model suggests that the harmonization of the education system across countries enhances student mobility between nations by reducing immigration costs. Employing a staggered difference-in-differences model, I find that on average, when two member countries have implemented the reform, the origin country experiences an increase in the outflow of students to the destination country by approximately 35.3%. These results provide compelling evidence that the implementation of a standardized tertiary education system successfully accomplishes the objective of enhancing student mobility, as outlined by the Bologna Process. The second chapter evaluates the effect of imposing tuition fees in Sweden on the location decision of international university students. Using a synthetic difference-in-differences model, I compare non-Swedish EEA students, who are not affected by the policy, to students from a set of developed countries who face this rise in tuition costs over a period of 17 years. I find that, on average, a decrease of 140 students per country of origin, which corresponds to approximately 98% of the average number of students from Non EEA developed countries studying in Sweden, and the effect is stable over time. Additionally, I find weak evidence that fewer foreign students impact the decision of Swedish students to study abroad in the short run
Estrogen Signaling in the Ventromedial Hypothalamus Modulates Adipose Tissue Metabolic Adaptation
Brain estrogen receptor α (ERα) signaling plays a vital role in regulating energy homeostasis and adipose tissue metabolism. It has been demonstrated that ERα is abundantly expressed in the ventrolateral region of the ventromedial hypothalamus (vlVMH), a sex-dimorphic structure that directly modulates brown adipose tissue (BAT) thermogenesis. We previously showed that an ERαvlVMH-originated neural circuit responds to changes in ambient temperature and nutritional states, suggesting a potential role of this neural circuit in metabolic adaptation. Here, we found that chronic activation of ERαvlVMH neurons prevents diet-induced obesity (DIO), which is associated with increased BAT thermogenesis and core temperature. Additionally, ERαvlVMH activation induces torpor-like phenotypes in fasted females. Conversely, chronic inhibition of ERαvlVMH neurons increases adiposity and decreases baseline BAT temperature. Notably, ERαvlVMH inhibition also impairs cold-induced food intake and BAT thermogenesis, resulting in lethal phenotypes during chronic cold exposure. Post hoc histology analysis revealed that ERαvlVMH inhibition induces adipose tissue whitening. Finally, we explored the anatomical distribution of ERαvlVMH downstream neural circuitries. Anterograde tracing revealed ERαvlVMH-originated projections to the medial preoptic area (MPOA), dorsal raphe nucleus (DRN), and substantia nigra (SNR). Interestingly, transsynaptic retrograde tracing further demonstrated that these brain regions send projections to inguinal white adipose tissue (iWAT), suggesting potential ERαvlVMH-originated circuits involved in adipose tissue adaptation. Chronic inhibition of ERαvlVMH→DRN circuit impaired cold-induced thermogenesis in females only, whereas chemogenetic activation of ERαvlVMH→MPOA circuit reduced thermogenic activity during cold exposure. These findings suggest that these circuits exert opposing effects and that ERαvlVMH-originated pathways regulate adipose tissue adaptation to environmental challenges in a sex-specific manner. Together, these findings support a model in which environmental challenges trigger ERαvlVMH activation to modulate fat-specific outputs and regulate adipose tissue adaptation to environmental challenges.
Live Library Upgrades via CRIU: A Novel Approach to Patch Security Vulnerabilities in Running Processes
Ensuring the security and reliability of long running software without service interruption is a persistent challenge. This thesis introduces a technique for live library upgrades that replaces vulnerable or outdated shared libraries in a running process without terminating it. Using Checkpoint/Restore in Userspace (CRIU), we freeze a process, modify its state, and resume execution with the patched code seamlessly integrated.
The approach combines metadata edits, address translation, and memory dump rewriting. By patching file backed mappings and leveraging ELF internals, we inject the new library while the dynamic loader transparently re resolves references. A synthetic execution captures memory resident data—such as the Global Offset Table (GOT)—from the updated library, and custom scripts adjust memory so the solution works under full Address Space Layout Randomization (ASLR).
Evaluation relies solely on publicly available open source software: we download multiple library versions, apply live upgrades, and run functional tests—no other datasets are used. Most notably, we live patch the compression library of a running SSH server with minimal downtime, after which the service continues operating correctly
By Community, with Community: Decolonial Recommendations for Collections Management Policies in Museums
This thesis is a study of the current collections management policies (CMP) that govern museums and that produce and sustain colonial language and practices. I focus on the impact of CMPs on community museums, and I conclude with recommendations for decolonial interventions, with the Filipino American Historical Society of Chicago (FAHSC) serving as the central case study in developing these solutions. My critical analysis of CMPsreveals the persistence of coloniality through the policies’ emphasis on permanent, singular ownership; the ratification of exclusionary professional standards; the use of vague operational language; and lack of community involvement. Thus, in order to be a decolonial institution, the Filipino American Historical Society of Chicago and other similar organizations should confront these issues through policies that center community agency, anti-accumulative practices, and multiple forms of knowledge production and sharing.
My research draws on critical policy/rhetorical analysis as the main methodology and decolonial as the main theoretical framework. I applied critical rhetorical analysis in my study of the collections policies of five community museums in the United States, and I conducted interviews with collections employees from these institutions to learn more about the development and enactment of such policies. To create concrete recommendations for FAHSC, this thesis explores the nuances of Filipinx/a/o identity formation and discusses the value of kapwa and kuwentuhan as lenses for framing such recommendations. In pursuing these solutions, this thesis desires to demonstrate the possibility of building collections policies that are built both by and for the community, as well as offer practical guidelines for local institutions to pursue their own decolonial collections work
Border Control Paradox: The Political Economy of Smuggling Between Colombia and Venezuela
This dissertation examines how smuggling economies are governed in contexts of contested sovereignty, focusing on the Colombia–Venezuela border between 2015 and 2022. Through extensive ethnographic fieldwork in Norte de Santander and adjacent Venezuelan territories, the study develops a theory of criminal governance in borderlands, analyzing how non-state actors—particularly the National Liberation Army (ELN)—exercise authority, regulate flows, and interact with state institutions. The research is guided by a central paradox: border control policies intended to suppress organized crime, such as militarization and border closures, often produce the opposite effect—empowering criminal actors, deepening violence, and institutionalizing hybrid forms of authority. This "border control paradox" is explored by tracing the evolution of the ELN from a rural insurgent group into a binational actor with territorial control over smuggling routes and informal crossings.
Methodologically, the dissertation combines political ethnography with documentary analysis, relying on interviews with smugglers, migrants, law enforcement officials, and local residents. The analysis reveals the emergence of plural governance arrangements, where state and criminal actors coexist, negotiate, and sometimes collude. It also documents the moral economies, everyday roles, and relational infrastructures that sustain the illicit trade of gasoline, food, and migrants. Rather than treating smuggling as a residual or deviant economy, the study situates it within a broader political economy of crisis, survival, and informal rule.
The findings contribute to debates on organized crime, state formation, and sovereignty by offering a grounded perspective on how borders are governed from below. They also hold policy relevance for designing security and humanitarian responses that avoid reinforcing criminal governance. Ultimately, the dissertation argues for a shift in how we conceptualize criminal power—not as external to the state, but as entangled with it, particularly in marginalized and strategic geographies like borderlands
The Earth is My Altar: A Black Trans Museology
In contending with slavery’s ubiquitous shadow over civil society and museum spaces, I revel in Blackness and transness’ ability to become a portal, or a disruption in the white supremacist, heteropatriarchal nature of the museum. As two social conditions defined by constructions that privilege whiteness and hetero and cis-normativity, Blackness and transness have resisted the fixity that positions them as singularly subjugated. This work posits that Black transness is an evocation of malleability, it is a doorway, gateway, and a series of linkages that propels possibility and resistance for Black trans people. Examining two instances in the history of museums in the United States, the 1893 World’s Columbian Exposition and Black Trans Liberation: Memoriam and Deliverance on view at the Modern Museum of Art PS1 in 2021, this thesis aims to illustrate museums’ status as perpetrators of social death, position Blackness and transness as disruptors of time and coloniality as represented in museum spaces, and ultimately, identify practices of remembrance that support Black trans flourishing
Making Meanings of the Body: What Cyborg Narratives Reveal About Prosthetic Limb Adoption and Use
In my project, I speak with 20 prosthesis-users, adapting timeline analysis alongside photo-elicitation and semi-structured interviews to examine participants’ cyborg narratives, the stories of how their use of these intimate technologies is integrated into their narratives of self. I ask two questions: RQ1) How do people who use prosthetic limbs relate to those objects as assistive technologies that have functional, aesthetic, and social features? RQ2) What do narratives of prosthetic limb adoption tell us about what these objects mean to their users? Through my analysis, I found that my participants view their prostheses primarily as tools whose use is determined by their ability to fulfill needs for efficiency and comfort. While users’ aesthetic desires vary widely, these preferences are important, enabling performances of gender and impacting one’s sense of self-presentation. Prosthesis use also facilitates community building and advocacy behaviors wherein prostheses act as conversation-starters with other amputees as well as the public and professional experts alike. By examining these aspects of prosthesis use through a narrative lens, one can see how these objects are meaningful to their users, not simply items that they use but intimate technologies that hold personal and social significance. While the photo-elicitation completed provided tangible grounding for the interviews as we discussed technological advancements and changes made between each prosthesis a participant used over their lives, the narrative approach further provided emotional resonance, helping underline not just the technicalities of technological changes but the salience and meaning of these changes in the lives of the participants. Thus, I argue for the continued application of visual and narrative approaches to exploring technology use over time to focus on the experiences of technology use and how technologies, especially assistive technologies, hold personal and social meaning to their users. Through this project, I highlight how the experiences and desires of users focus on the importance of a neutral experience (rather than one focused on going “above and beyond” expectations), and how these goals differ from those expected by non-amputees. This underlines the importance of centering user perspectives in the design of all assistive technologies
Scalable and Tunable Algorithms for Adaptive All-to-all Data Exchange
All-to-all data exchange, involving bidirectional data transfer among all participating processes, is a fundamental class of data movement in HPC systems. This contains two modalities: uniform (fixed-size) and non-uniform (variable-size) exchange patterns. The significance of these exchange protocols has grown considerably with the advent of exascale computing, as they serve as critical communication primitives for numerous large-scale applications. These include data-driven machine learning, graph mining, fast Fourier transform, quantum computer simulations, and certain advanced preconditioners and solvers. Implementing all-to-all in a manner that scales for different workload sizes, distribution patterns, process counts, and hardware configurations is very challenging.
To facilitate scalability across various configurations, tunability emerges as the principal consideration. The ability to dynamically tune execution parameters for maximum bandwidth utilization while minimizing latency overhead across different input configurations is key to achieving scalability. This requires an analytical focus on two critical metrics: the total number of communication rounds (latency-relevant) and the total volume of data exchanged (bandwidth-relevant). Existing implementations typically rely on algorithms with extremal radix (log-base) values of 2 (latency minimization) and process count (bandwidth maximization). This limits the ability to dynamically optimize performance across the latency–bandwidth spectrum and significantly constrains adaptability to diverse demands. Additionally, current approaches illustrate a fundamental gap between theoretical algorithm design and practical hardware constraints, forfeiting performance gains available through multi-tiered parallelism in modern distributed systems.
This thesis addresses these challenges by designing scalable and tunable algorithms for adaptive all-to-all data exchange, enabling optimal performance across various communication demands. To achieve it, we first formulate a mathematical model that characterizes communication patterns in uniform all-to-all through parameterized expressions quantifying total communication rounds and data exchanged. This model facilitates informed radix selection and effectively balances latency and bandwidth considerations. Subsequently, we develop a tunable radix-based algorithm for uniform all-to-all, which is then extended to non-uniform cases via a novel two-phase communication scheme. We then design hierarchical parameterized algorithms for both uniform and non-uniform all-to-all, strategically leveraging the multi-tiered parallelism of modern HPC architectures. Lastly, we integrate machine learning techniques to autonomously calibrate algorithmic parameters, enabling dynamic and adaptive all-to-all optimization
Long-time Behavior and Multiscale Description of Models of Collective Dynamics
We study the well-posedness and long-time dynamics of various models of collective dynamics
The Impact of Early Life Adversity on Neural Reward Reactivity in Early Childhood
Early Life Adversity (ELA), defined as childhood experiences which require significant adaptation beyond the child’s developmental stage, is of critical concern, as it is predictive of increased risk for the leading causes of morbidity and increased severity, chronicity, and overall risk for psychiatric disorders. Dysfunction in neural reward systems may represent a mechanism through which ELA increases risk for later psychopathology; however, few studies have examined the proximal impact of ELA on reward processing in early childhood, prior to periods of enhanced risk for psychopathology onset. Conceptual models suggest that different forms of ELA may differentially impact neural reactivity, and indeed, studies find that exposure to deprivation, such as neglect, may be more strongly tied to blunted reward processing than experiences of threat to one’s safety. The reward positivity (RewP) is an event-related potential component which indexes neural reward responsiveness. Preliminary evidence indicates support for dysregulated processing styles (i.e., reduced RewP) related to greater ELA exposure; however, this relationship has been relatively unexamined in children. In addition, decomposition of the RewP into frequency (i.e., delta, theta) rather than time-domain represents an innovative, novel method for enhancing nuanced understanding of the targeted influence of ELA on neuroaffective processing. The current study examined relations between ELA composites, including threat and deprivation dimensions, with RewP and underlying delta activity from 350-425 at a pooling of frontal, central, and parietal electrodes in 80 5–6-year-old children, half of which were risk-enriched through maternal depression history. Linear regressions examining the relation between ELA composites and RewP and underlying delta demonstrated that ELA composites were not significantly related to neural reward processing.
Moderation by parenting styles was explored, which revealed that parenting styles did not exert interactive effects with ELA on reward processing. In exploratory analyses, current maternal depression symptoms moderated the relation between deprivation and RewP, such that youth with higher maternal depression symptoms and greater deprivation exposure demonstrated more blunted reward reactivity. Taken together, these results suggest that ELA composites alone are not significantly related to reward reactivity in early childhood. However, maternal depression symptoms may be among several factors that moderate ELA and reward reactivity relations, indicating a subgroup of youth at enhanced risk for reward processing alterations in association with adverse childhood experiences. Further research could help to identify youth at highest risk for reward processing alterations and to inform which youth may benefit most from interventions targeting enhancing positive valence system functioning