University of Illinois Urbana-Champaign
Illinois Digital Environment for Access to Learning and Scholarship RepositoryNot a member yet
123813 research outputs found
Sort by
Network cerebrovascular health predicts cognitive control performance
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo termsThe student, Xiang Quan, accepted the attached license on 2025-03-12 at 19:59.The student, Xiang Quan, submitted this Thesis for approval on 2025-03-12 at 20:33.This Thesis was approved for publication on 2025-03-24 at 11:11.DSpace SAF Submission Ingestion Package generated from Vireo submission #21673 on 2025-10-19 at 18:09:09Cerebrovascular health has emerged as a crucial factor in influencing cognitive control performance. Previously, most studies have relied on basic measures of cerebrovascular health due to the lack of precise regional estimates. This study utilizes a novel pulse-DOT method to provide detailed regional Pulse Relaxation Function (PReFx) measurements, offering valuable insights into regional vascular elasticity. For the first time, we integrate a network-based approach from fMRI studies to cerebrovascular health research, as the support provided by the cerebrovascular system to different brain networks may differentially impact cognitive control performance. To comprehensively assess how various mechanisms of cognitive control are influenced by network cerebrovascular health, we employed a task-switching paradigm that simultaneously accesses proactive and reactive control. Our findings reveal that larger positive deviations in PReFx for the Dorsal Attention Network and the Visual Network from global PReFx measures are linked to reduced cognitive control costs. Conversely, larger positive deviations in PReFx for the Default Mode Network and Fronto-Parietal Network from global PReFx measures are associated with increased cognitive control costs. These results indicate that the relative balance in cerebrovascular function across different brain networks impacts cognitive control, highlighting the value of studying network-based cerebrovascular status in understanding cognitive performance
Effects of yeast products and dietary fibers on gastrointestinal health outcomes and microbiota populations of dogs and cats
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo termsThe student, Vanessa De La Guardia Hidrogo, accepted the attached license on 2025-03-19 at 12:28.The student, Vanessa De La Guardia Hidrogo, submitted this Dissertation for approval on 2025-03-19 at 13:58.This Dissertation was approved for publication on 2025-03-20 at 10:28.DSpace SAF Submission Ingestion Package generated from Vireo submission #21685 on 2025-10-19 at 18:09:11Over the past few decades, interest in functional ingredients has increased, leading to their broad application in pet foods. This trend is largely driven by the growing commitment of owners to their pets' health and well-being. Functional ingredients are incorporated into pet diets to provide health benefits that extend beyond basic nutrition, supporting areas such as gastrointestinal health and immune function. Given the high costs and time constraints associated with traditional animal experiments, the use of both in vivo and in vitro models offers a more efficient approach for the preliminary evaluation of the functional benefits of these ingredients. The objective of this dissertation was to evaluate the effects of select functional ingredients on gastrointestinal health outcomes and microbiota populations of dogs and cats. The aim of the first experiment was to evaluate the effects of a yeast-enriched functionalized canola meal (FCM) on apparent total tract digestibility (ATTD) of diets and their effects on the fecal quality, metabolite concentrations, and microbiota populations, and immune function of healthy adult dogs. Twelve adult female beagle dogs (body weight = 7.6 ± 0.7 kg; age = 5.8 ± 1.3) were used in a replicated 4 × 4 Latin square design to test the following dietary treatments: 1) FCM only (no yeast inclusion), 2) FCM + low yeast dose, 3) FCM + medium yeast dose, and 4) FCM + high yeast dose. All treatments were top-dressed onto the basal diet at a rate estimated to be 1% of daily intake (as-is basis). Supplementation with yeast-enriched FCM had no effect on the ATTD of macronutrients or energy or the fecal characteristics, metabolite concentrations, and microbiota populations or immune function of healthy adult dogs. The aim of the second experiment was to evaluate the effects of extruded diets containing FCM enriched with different levels of dried yeast on the ATTD of diets and their effects on fecal characteristics and microbiota, skin measures, and serum oxidative stress markers of adult dogs Twelve adult female beagle dogs (body weight = 9.75 ± 0.83 kg; age = 6.2 ± 1.6 yr) were used in a replicated 4 × 4 Latin square design to test four extruded kibble diets: 1) CTRL (containing no FCM or yeast), 2) FCM + low yeast dose (LY), 3) FCM + medium yeast dose (MY), and 4) FCM + high yeast dose (HY). Supplementation with yeast-enriched FCM tended to increase fecal output (as-is: P 1%) bacterial genera were affected by significant fiber×time interactions. Specifically, acacia fiber tubes had a greater (P<0.01) increase in Bacteroides, Blautia, and Faecalibacterium than other tubes. Inulin and pectin tubes also had a greater increase (P<0.01) in SCFA-producing bacteria. Our findings suggest that the effects of yeast-enriched FCM are largely dependent on the ratio of yeast to FCM, with larger effects observed at greater yeast inclusion levels. Yeast-enriched FCM may positively influence stool quality, fecal metabolites, and microbiota when incorporated into extruded diets. Additionally, the in vitro experiments demonstrated the influence of dried brewer's yeast and various fiber sources on microbial composition and fermentation dynamics. Fiber fermentation profiles vary widely and can significantly affect canine and feline microbiota community composition and activity. Acacia fiber is moderately fermentable, promoting bacterial diversity and beneficial shifts in microbiota composition and activity while producing minimal gas. In contrast, pectin and inulin exhibit high fermentability, with significant effects on gut microbiota populations and high gas production, whereas cellulose is poorly fermentable resulting in minimal effects on microbial composition and activity. Thus, inclusion of yeast-based ingredients and moderately fermentable fibers in pet foods may promote gastrointestinal health
Regression test prioritization for modern software
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo termsThe student, Runxiang Cheng, accepted the attached license on 2025-04-15 at 19:21.The student, Runxiang Cheng, submitted this Dissertation for approval on 2025-04-15 at 19:35.This Dissertation was approved for publication on 2025-04-16 at 09:38.DSpace SAF Submission Ingestion Package generated from Vireo submission #21786 on 2025-10-19 at 18:09:24Continuous Integration (CI) is a common practice in software development for increasing code quality. CI runs regression test suite on each code change to help software developers find faults in the change. As codebase size and code change frequency have grown rapidly in modern software development, regression test suite runtime has also increased, thereby preventing timely debugging feedback to developers. Regression Test Prioritization (RTP) aims to expose faults in code change sooner by reordering tests in the test suite, so that the ones likely to fail are run earlier. Despite that RTP has been extensively studied for nearly three decades, research results of RTP are rarely put into practice. To facilitate the adoption of RTP for modern software systems, RTP must be demonstrated effective in speeding up regression fault detection in relevant and practical testing settings. Moreover, robust and efficient RTP tooling must be available for practitioners to easily apply RTP research results in popular software ecosystems. This dissertation takes a step forward in addressing these challenges via three lines of work. First, this dissertation examines key research findings from prior RTP studies on a newly-proposed dataset of long-running test suites with real test failures and up-to-date CI practices. Through extensive evaluation, this dissertation identifies the most effective and efficient RTP techniques under the impact of realistic CI issues such as flaky test failures. Second, this dissertation applies traditional RTP techniques for configuration testing, an emerging but critical testing scenario; it further proposes novel and effective RTP techniques for configuration testing. This dissertation assesses various factors that influence the effectiveness of RTP in speeding up misconfiguration detection, and provides detailed analyses and guidelines in applying RTP to configuration tests. Third, this dissertation presents pytest-ranking, a readily-usable open-source RTP tool for Python and its most popular testing framework Pytest. This dissertation realistically evaluates the efficacy of pytest-ranking, demonstrating that pytest-ranking is well integrated with Python/Pytest ecosystems, can be conveniently deployed into CI, has low overhead, and can detect faults substantially faster than baselines
Learning symbolic concepts and domain-specific languages
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo termsThe student, Paul Krogmeier, accepted the attached license on 2025-04-19 at 10:27.The student, Paul Krogmeier, submitted this Dissertation for approval on 2025-04-19 at 11:00.This Dissertation was approved for publication on 2025-04-21 at 11:37.DSpace SAF Submission Ingestion Package generated from Vireo submission #21846 on 2025-10-19 at 18:09:33Symbolic concept learning is a fundamental problem with numerous applications in program testing, verification, synthesis, and discovery of mathematical theorems and physical laws. We investigate in this dissertation the decidability of symbolic learning in different formal languages and provide new decision procedures for their learning problems. We develop an algorithm based on tree automata that can decide the existence of logic formulas which correctly classify a given set of labeled structures, where formulas come from fragments of finite-variable first-order logics defined by regular constraints on formula syntax trees. We show that alternating tree automata can evaluate quantified logic formulas over finite structures and use this to give an exponential time upper bound for learning. We prove a matching lower bound. We show also that finite-variable logics with recursion have decidable learning by moving to two-way tree automata which evaluate recursive definitions by traversing syntax trees up and down several times. We also prove decidable learning results for finite-variable second-order logics and Datalog programs. Next we show that these learning algorithms generalize to languages beyond finite-variable logics. We prove a meta-theorem which says that learning in a language L is decidable if we can devise a restricted program P which evaluates expressions of L over fixed structures. Provided that P does not need more and more memory for larger and larger expressions, it can be translated with a learning instance into a tree automaton that accepts exactly the set of solutions; this reduces learning problems to the emptiness problem for these tree automata. We use the meta-theorem to prove new decidable learning results by exhibiting evaluation programs for several languages: modal logic and computation tree logic over finite Kripke structures, linear temporal logic over ultimately periodic words, regular expressions and context-free grammars over finite words, first-order logic queries over rational numbers, and text-manipulation programs for spreadsheet automation. We next turn to the problem of synthesizing domain-specific languages (DSLs) which succinctly express symbolic concepts in specific domains and thereby enable learning from few examples. We formulate novel DSL synthesis problems involving the synthesis of grammars with and without macros, and we prove decidability results for each. Finally, we formulate an axiom synthesis problem in which the goal is to synthesize a set of logic formulas which precisely characterize a target class of mathematical structures. We propose a general learning-based algorithmic framework for solving axiom synthesis and instantiate it to synthesize axiomatizations in modal logic and Kleene algebra
Waferscale network switches
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo termsThe student, Shuangliang (David) Chen, accepted the attached license on 2025-04-21 at 14:02.The student, Shuangliang (David) Chen, submitted this Thesis for approval on 2025-04-21 at 14:10.This Thesis was approved for publication on 2025-04-21 at 16:19.DSpace SAF Submission Ingestion Package generated from Vireo submission #21861 on 2025-10-19 at 18:09:35In spite of being a key determinant of latency, cost, power, space, and capability of modern computer systems, network switch radix has not seen much growth over the years, due to poor scaling of off-chip IO pitches and switch die sizes. We consider waferscale integration (WSI) as a way to increase the size of the switch substrate to be much bigger than a single die and ask the question: can we use WSI to enable network switches that have dramatically higher radix than today’s switches? We show that while a waferscale network switch can support up to 32x higher radix than state-of-the-art network switches when only area constraints are considered, the actual radix of a waferscale network switch is not area-limited. Rather, it is limited by a combination of internal bandwidth, external bandwidth, and power density. In fact, without optimizations, benefits of a waferscale network switch are minimal. To address the scalability bottlenecks, we propose a heterogeneous network switch design that reduces switch power by 30.8%-33.5% which, in turn, allows an increase in radix (by up to 4x) by increasing internal I/O bandwidth at the expense of energy efficiency. We also propose subswitch deradixing that increases the overall radix by 2x by decreasing the radix of the subswitches to alleviate the internal I/O bottleneck. We use Area I/O and Optical I/O schemes to alleviate the external I/O bandwidth bottlenecks of conventional SerDes-based external connectivity. In addition to scalability optimization, we present optimizations such as low latency buffering and proprietary routing that improve the performance of waferscale switches. Finally, we present a system architecture for a waferscale network switch that supports its port count, power delivery, and cooling requirements in a compact form factor. We show that the switch can be used to enable new computing systems such as single-switch datacenters and massive-scale singular GPUs. It can also lead to a dramatic reduction in datacenter network costs. Overall, this is the first work quantifying the benefits of waferscale switches and identifying and addressing the unique challenges and opportunities in building them
Physiological effects of enhanced weathering with basalt on maize and soybean grown under increased carbon dioxide
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo termsThe student, Megan Allen, accepted the attached license on 2025-04-21 at 17:09.The student, Megan Allen, submitted this Thesis for approval on 2025-04-21 at 17:22.This Thesis was approved for publication on 2025-05-05 at 15:13.DSpace SAF Submission Ingestion Package generated from Vireo submission #21873 on 2025-10-19 at 18:09:38Enhanced weathering (EW) is a carbon dioxide (CO2) removal technology in which silicate rocks, such as basalt, are finely ground and applied to land to accelerate natural weathering processes that capture and store atmospheric CO2 securely in the ocean. Croplands have been targeted for EW schemes because of the potential agricultural co-benefits of silicate powders, including improved soil pH and nutrient provision. The Midwestern United States could be an ideal region for integrating EW due to its large cropland area, mostly consisting of maize and soybean, and its existing infrastructure. If EW becomes more common place, it is important to consider factors that might affect the system in the future. If global CO2 emissions are not drastically cut, atmospheric CO2 concentrations will continue to rise, with potential implications for both crop production and EW rates. This study investigates how silicate powders might affect maize and soybean physiology and how changes in crop physiology might affect EW in a higher CO2 future. Aboveground physiological responses as well as soil moisture and respiration were measured over two growing seasons under ambient and increased CO2 levels in a fully open-air field setting. The application of basalt (40 t ha-1 year-1) did not significantly alter any aboveground physiological responses in maize or soybean under ambient or increased CO2 levels. This lack of response was likely due to the site’s relatively neutral starting pH, which reduced the potential of basalt to improve nutrient availability. In maize, increased CO2 enhanced photosynthesis during an early to mid-season drought, aligning with previous research showing that C4 crops experience CO2-related benefits only under drought stress. However, this did not translate to increased grain yield. In soybean, increased CO2 also did not improve seed yield, likely because a mid-season drought reduced the typical improvements to C3 photosynthesis observed under increased CO2. Increased CO2 also affected soil moisture at 10 – 50 cm depth in both cropping systems, with improvements to soil water only observed during non-drought periods. Basalt increased soil moisture at 40 – 50 cm across both crops, likely due to altered root dynamics, with a smaller effect under elevated CO2 in maize. In soybean, basalt also improved topsoil moisture, potentially through improvements to water-holding capacity, although further research is needed to confirm this. Neither basalt nor increased CO2 affected soil respiration rates. These findings show little evidence for interactions between basalt and increased CO2 on aboveground and belowground crop responses. It is possible, however, that under non-drought conditions, increased CO2 could improve Midwestern EW rates via greater soil moisture, although direct measures of EW were outside the scope of this study. This study also highlights the importance of initial soil conditions when assuming that basalt will stimulate crop productivity
Occlusion-aware tracking for drones using neural methods
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo termsThe student, Akul Datta, accepted the attached license on 2025-04-24 at 18:43.The student, Akul Datta, submitted this Thesis for approval on 2025-04-24 at 18:47.This Thesis was approved for publication on 2025-04-25 at 10:55.DSpace SAF Submission Ingestion Package generated from Vireo submission #21967 on 2025-10-19 at 18:10:56This thesis addresses the challenge of gate tracking under occlusion conditions, which remains a significant hurdle for robust autonomous systems. Occlusions occur when tracked objects become partially or completely hidden behind other objects or leave the camera’s field of view, which can cause conventional trackers to lose target identity and accurate positioning. We present a novel approach that combines synthetic data generation and specialized neural network architectures to maintain tracking performance even when the target is partially or completely hidden from view. We implement and experimentally compare four distinct approaches: LiteTracker-CNN with a ResNet18 backbone and specialized point regression for gate corner tracking, LiteTracker- ViT using cross-frame attention to maintain spatial coherence during occlusions, Fine-Tuned CoTracker3 (FT) adapted using synthetic AirSim data, and a semi-supervised fine-tuned CoTracker3 (FT Pseudo) with pseudo-labeling for unlabeled footage. Using Microsoft Air- Sim simulator-generated datasets containing precise ground truth annotations even during occlusion events, we systematically evaluate each approach’s performance across controlled occlusion scenarios. Our experimental results, validated through a benchmark across different occlusion types (non occluded, partially occluded, and fully occluded), reveal that each architecture excels in different aspects: ViT provides the most robust tracking through occlusion events, CNN delivers the fastest processing for resource-constrained applications, and FT offers a balanced middle-ground solution. The ViT implementation specifically achieves a nearly 99% success rate on occluded points and 82% lower error metrics compared to baselines. Analysis of performance degradation showed that ViT experiences only 35.7% error increase post-occlusion compared to CNN’s 47.4% and baseline models’ 540+% degradation. Our supervised fine-tuning approach significantly outperformed the pseudo-label self-supervised method for occlusion handling, with a success rate of 97.11% compared to 46.22% with the pseudo-label approach. This demonstrates that even limited annotated data substantially improves tracking performance through occlusions
Traumatic migration experiences and the development of PTSD in Latino youth: A scoping review
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo termsThe student, Gabriella Lopez, accepted the attached license on 2025-04-29 at 10:16.The student, Gabriella Lopez, submitted this Thesis for approval on 2025-04-29 at 10:17.This Thesis was approved for publication on 2025-05-05 at 13:05.DSpace SAF Submission Ingestion Package generated from Vireo submission #22070 on 2025-10-19 at 18:11:14Latin America is a region that has been plagued by civil unrest, state corruption, gang violence, and economic instability for decades. These circumstances have influenced or forced many individuals, including children and adolescents, to journey north. However, the experience is often filled with its own kinds of dangers: robbery, kidnappings, extortion, physical assault, sexual assault, human trafficking, environmental hazards, and dangerous transportation methods are frequent occurrences. Exposure to these dangerous and violent encounters may lead to the development of mental illnesses such as post-traumatic stress disorder (PTSD), especially in vulnerable populations like youths. There is substantial evidence for pre-migration trauma and mental disorders in Latinos. However, there is a gap regarding migration trauma and the potential onset of PTSD, as well as how children and adolescents may be impacted by trauma exposure during migration. The purpose of this scoping review was to explore the literature to discover if traumatic migration experiences led to the development of PTSD in Latino youths and if rates differed when youths were accompanied or unaccompanied. The term “Latine”, a gender-neutral term that has roots in Spanish-speaking countries, will be used throughout the review alongside “Latino”. PubMed, PsychInfo, PsycArticles, and Google Scholar were used to extract articles. Studies were eligible for inclusion if the population of interest were Latino immigrant children or adolescents, the exposure was migration trauma experienced or witnessed, and the outcome was PTSD. Articles were excluded if they focused on adults, non-migration trauma including pre- or post-migration, or did not focus on PTSD as an outcome. After duplicates were removed, 202 articles were left to screen. Only three articles were eligible. Significant correlations were observed between trauma exposure during migration and PTSD. Specifically, physical assault exposure and child physical abuse during migration were found to significantly predict PTSD symptoms. No differences in PTSD rates were reported between youths who traveled accompanied versus unaccompanied. Youths who were exposed to traumatic events during migration are at an increased risk of developing PTSD. It should be noted that there was a dramatic gap in the literature on migration trauma exposure and PTSD outcomes in Latine youths. More funding and research are needed to understand the complexity of trauma exposure, migration, and mental health in Latino youths to better provide and support these individuals once they arrive in the U.S
Combustion-ablation time scale phenomenology
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo termsThe student, Julia Brown, accepted the attached license on 2025-04-29 at 21:52.The student, Julia Brown, submitted this Thesis for approval on 2025-04-29 at 21:54.This Thesis was approved for publication on 2025-05-05 at 15:07.DSpace SAF Submission Ingestion Package generated from Vireo submission #22101 on 2025-10-19 at 18:11:22The thermal protective systems designed for atmospheric reentry are analyzed through sub-mechanisms to understand the multiphysics of integrated structures and inform design choices for scramjet combustors and material optimization. Through the interaction of different mechanisms acting on different time scales, the phenomenology for material degradation of carbon fiber ablators is studied numerically. A simple one-dimensional model was an appropriate first step for analysis of the thermal conduction and oxidation of the material, with nonlinear implicit methods, that facilitate solutions with widely disparate time scales. The numerical solution was verified through a method of manufactured solutions. The model was then expanded, coupling additional physical mechanisms into the system including mass conservation, oxygen transport, and advective velocity within the porous media, thus creating a thermal boundary layer, limiting material oxidation. The different time regimes allow for the comparison of non-dimensional ratios that produce drastically different behaviors for the system. By developing an understanding of multiphysics interactions with time scales for the combustor, optimized design and material constraints can be quickly and affordably determined for experimentation
Surface actuation based flow control for low Reynolds number flow past an airfoil
No full textSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo termsThe student, Ernold Thompson, accepted the attached license on 2025-04-30 at 20:53.The student, Ernold Thompson, submitted this Thesis for approval on 2025-04-30 at 20:58.This Thesis was approved for publication on 2025-05-05 at 13:26.DSpace SAF Submission Ingestion Package generated from Vireo submission #22136 on 2025-10-19 at 18:11:25In this thesis, surface actuation for the flow past an airfoil at a Reynolds number of 1000 has been studied numerically. First, prescribed normal actuation in the form of a backward traveling wave on the suction surface with sinusoidal spatial and temporal characteristics has been studied for the purpose of lift improvement. The kinematics of actuation are defined by wavenumber and wavespeed, both of which are varied over a wide range of values to include parameters that considerably change the lift dynamics as well as those that do not. First, the effect of actuation at an angle of attack of α = 5◦ where the unactuated flow is steady is considered. Lift benefits are found to be maximal when the morphing kinematics align with the intrinsic advection velocity of the flow. Then, the role of morphing in the presence of an unsteady, separated baseline flow (with intrinsic vortex-shedding processes) at α = 15◦ is investigated. At this higher angle of attack, three distinct behavioral regimes based on the relationship between morphing and the underlying shedding frequency are identified. Actuation is found to yield lift benefits when the actuation time scales are close to those of shedding with the highest increase in mean lift occurring when the actuation frequency is slightly smaller than the shedding frequency of the unactuated flow. The influence of the actuation on the flow field is discussed in terms of the variations in the pressure field resulting from the velocity boundary condition imposed by surface actuation. The velocity boundary condition imposed by actuation is also used to describe the variations in vortex shedding behavior at α = 15◦. Although backward traveling wave on the suction surface can lead to improvements in mean lift, a concurrent increase in drag occurs. To explore actuation for the separate purposes of lift improvement and drag mitigation, in the second part of this thesis numerical optimization is used to determine the actuation for the separate goals of lift improvement and drag mitigation at α = 15◦ with actuation permissible on the entire surface of the airfoil (including the pressure side). The gradient information required by the optimization algorithm is computed in a computationally efficient way using the adjoint of the governing equations. The optimal actuation profiles for the two performance aims are compared. Where possible, similarities with actuation in the form of backward traveling waves on the suction surface have been highlighted. The flow features emerging from the optimal actuation variations, and their consequent influence on the instantaneous aerodynamic coefficients, have been analyzed