Grants

NSF

Understanding the internal structure of protons and neutrons—collectively called nucleons—is a long-standing challenge in nuclear physics. This project seeks to create a three-dimensional map of how quarks and gluons, the fundamental constituents of matter, move and interact inside nucleons. By analyzing data from high-precision experiments at Jefferson Lab and preparing for measurements at the future Electron-Ion Collider (EIC), the research will investigate how quark and gluon spin and motion contribute to nucleon structure. To interpret these complex datasets, the project applies advanced artificial intelligence (AI) tools—specifically deep learning and statistical methods—to improve measurement accuracy and quantify uncertainty. Beyond advancing fundamental knowledge, the project has broad societal impact. In alignment with the Nuclear Science Advisory Committee’s recommendation to integrate AI in nuclear research, it will train students in physics and data science, develop reusable tools for analyzing complex data, and generate accessible educational resources through workshops and tutorials. These efforts will help cultivate the next generation of scientific leaders. Notably, AI methods developed by the team for nuclear physics have also been applied for other data-intensive applications, demonstrating their versatility and broad relevance. This project advances national priorities in nuclear science and AI while fostering discovery and education. The project aims to study the three-dimensional partonic structure of nucleons through the analysis of semi-inclusive deep inelastic scattering (SIDIS) data collected with the CLAS12 detector at Jefferson Lab, focusing on kaon electro-production from unpolarized and polarized targets. By detecting final-state hadrons alongside the scattered lepton, SIDIS provides sensitivity to the transverse momentum of struck quarks. The central objective is to extract transverse momentum-dependent parton distributions (TMDs), which extend conventional parton distribution functions by incorporating transverse momentum in addition to longitudinal momentum fraction. These observables enable studies of spin-orbit correlations, such as the Boer-Mulders effect, offering insight into how quarks and gluons contribute to nucleon spin. The analysis will leverage two advanced deep learning tools developed by the PI’s group: ELUQuant, for event-level uncertainty quantification, and Deep(er)RICH, for reconstructing images from Cherenkov detectors and improving particle identification. Bayesian inference will be employed for robust extraction of physics observables, complementing traditional unfolding methods. The team will also conduct sensitivity studies for the ePIC detector at the future EIC. This project will advance multidimensional nucleon tomography, addressing key questions such as the proton spin puzzle and the role of orbital angular momentum. It supports both precision nuclear physics and AI-driven workforce development. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

With the support of the Chemistry of Life Processes program in the Division of Chemistry, Ling Hao from George Washington University is developing novel bioanalytical chemistry methods to study mitochondrial dynamics in human neurons. Mitochondria are the powerhouses of the cell; they produce ATP and function as hubs for many metabolic pathways. Neurons in the brain rely heavily on mitochondrial functions because have extremely high energy demands to support synaptic activities. However, the specific molecular microenvironment of neuronal mitochondria is not fully understood, partly due to technological limitations. This project will develop mass spectrometry (MS)-based techniques to characterize the dynamic mitochondrial activities and interactions at the molecular level in human, stem cell-derived neurons. Through partnership with local chemistry instrument companies and MS discussion group, a summer MS workshop will be organized to bring together students from different levels as well as scientists from local academic, industry, and government institutions. A set of research, education and outreach activities will be integrated to enhance student learning and training experiences at the university and regional levels and to promote accessibility and diversity in STEM (science, technology, engineering and mathematics). The proposed research aims to address major technical challenges and knowledge gaps to study molecular networks and the subcellular microenvironment in human neurons using mass spectrometry (MS) techniques. Despite the advancements in understanding mitochondrial functions, it is technically challenging to capture the dynamic mitochondrial microenvironment, particularly in neurons with highly polarized structures. A set of MS-based proximity labeling strategies will be developed to study mitochondrial activities, functions, metabolic regulations, and membrane dynamics in human induced pluripotent stem cells (iPSCs)-derived neurons. This research program will provide widely applicable chemistry tools to study sub-cellular biological microenvironment as well as provide new insights into mitochondrial dynamics and neurobiology. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

In this CAREER project, funded by the Chemical Mechanism, Function, and Properties Program of the Chemistry Division, Professor Pier Alexandre Champagne of the Department of Chemistry and Environmental Science at the New Jersey Institute of Technology is employing a combination of experimental and computational techniques to advance the chemical understanding of reactive sulfur species, which are of significant importance in biochemistry. The goal of this research is to access new synthetic donors of reactive sulfur species, use the donors to probe the behavior of those species, and apply computational tools to obtain a holistic understanding of their chemical reactivity. Regarding educational activities, the Visualize Organic Chemistry online learning platform will be expanded with new animations and tutorials that support greater mechanistic literacy among organic chemistry students and practitioners across the United States. Reactive sulfur species are a wide class of biochemical intermediates that have important physiological effects but that are still not well understood due to their thermodynamic and kinetic instability. The proposed computations will address this issue by characterizing the intrinsic reactivity of polysulfides with biological nucleophiles and electrophiles using Density Functional Theory calculations, allowing a holistic understanding of their behavior in vivo. The proposed synthetic donors of reactive sulfur species will provide access to new photoactivated tools to study the chemical reactivity and biochemical effects of reactive sulfur species, by leveraging the BODIPY photocage scaffold. These will enable the study of structural effects on persulfide reactivity, and the experimental characterization of the chemical properties of hydrotrisulfides, heralding a new era in the study of polysulfides that are critical to biochemistry. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

This CAREER project seeks to provide a comprehensive understanding of the sources, spatial concentrations, chemical processes, and climate impacts of micro- and nano- plastic particles in the atmosphere. Atmospheric micro- and nano- plastic particles contribute significantly to the mass of marine and land micro- and nano- plastic particles, and their concentrations are expected to increase ten-fold over the next decade. Their presence has significant environmental implications, but their atmospheric behavior and impacts remain poorly understood. The research will combine laboratory studies with mobile field measurements in the Houston Texas region to gain a holistic understanding of micro- and nano- plastic particles in the atmosphere. This research on micro- and nano- plastic particles (MNPPs) has three objectives: (1) the real-time and offline quantification of major atmospheric MNPPs to comprehensively examine their atmospheric concentrations, spatial distribution, and sources; (2) the characterization of atmospheric reaction kinetics, chemical reactivity, and decomposition products of MNPPs through laboratory oxidation studies; and (3) the investigation of the potential climate impacts of MNPPs by assessing their optical properties and cloud condensation nuclei (CCN) activities during their atmospheric lifecycle. This research includes the development of an interactive exhibition at the Brazos Valley Museum of Natural History to inform the local community about air pollution and atmospheric science, and the design of an interdisciplinary chemistry lesson module for a local high school. Both graduate and undergraduate students will participate in the research over the course of the project. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

One of the most important applications of fluid dynamics is improving the design of turbomachinery. These devices exchange energy between a fluid and a rotor. They are used in many applications such as marine propulsion, aircraft, and ocean energy technologies. They generate complicated turbulent flows that are difficult to analyze. On top of that, turbulent flows in the atmosphere or ocean interact with the flow generated by the turbomachinery, which is often ignored in engineering analysis. This project will develop new computational tools to analyze turbomachinery flows in realistic operating environments. It will generate accurate models for modern turbines and uncover effects of environmental flows on turbine performance. Results will be checked against wind tunnel and field data. The models generated in the project will be open-source and shared with the public on GitHub. The project will also develop hands-on activities for K-12 students to learn about wind and power generation. The outcomes of the project will help improve U.S. energy infrastructure and encourage students to pursue careers in energy engineering. This project uses scale-resolving large eddy simulations to systematically unravel nonlinear interactions between stratified boundary layer flows, rotor aerodynamics, turbine wakes (momentum and turbulence), and array/boundary layer-scale entrainment and wakes. To guide design and control optimization with accurate predictions, this project develops open-source fast engineering models of these coupled, multi-scale dynamics. The models will be developed using targeted decompositions that first parse rotor aerodynamics from wakes, enabling thrust and power predictions using first principles, and then isolate turbulent wakes from the background stratified, turbulent boundary layer. This project will validate the models using large eddy simulations, wind tunnel experimental data, and field data. The first objective uses scale-resolving large eddy simulations to elucidate the coupled impacts of rotor operation and velocity shear on rotor thrust and power. This project then models synergistic rotor and boundary layer effects on turbulent wakes and the large-scale environmental response to arrays of energy harvesting devices, resulting in an engineering model that couples the predictions of turbine-scale wakes and array-scale entrainment and yields lower error than existing models that neglect or parameterize shear, stability, and turbulence. Finally, the model will be extended, and validated using large eddy simulations, to predict large-scale wakes of arrays. Together, this project produces cross-cutting impact for applications including stratified turbulence and high Reynolds number geophysical flows. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

This project examines aerosol impacts on multivariate climate hazards that have great significance to society, including the concurrent extreme events of fire weather, humid-heat, and drought events. Aerosols have a larger impact on univariate extreme temperature and moisture events per unit global mean temperature change than greenhouse gases. However, their role in multivariate climate hazards is unknown and a framework on how to consider aerosols when attributing and planning for these hazards is also lacking. At the same time, aerosols are increasing because of increases in both anthropogenic and natural sources, accelerating the need to understand their role in climate hazards. This study addresses this need by 1) developing methodologies to uncover the influence of aerosols on multivariate climate extremes using state-of-the-art climate models; and 2) educating the next generation climate workforce with greater awareness of the factors contributing to multivariate climate extremes and the tools to enable robust societal planning for climate risk, taking into account these impacts. A key hypothesis of this work is that aerosols can have a greater impact than greenhouse gases on temperature and moisture and, thus, multivariate climate hazards in certain regions and for specific events. The project will examine this hypothesis using existing and emerging model output from the multi-model single forcing large ensemble (SFLE) aerosol intercomparison project and the Regional Aerosol Model Intercomparison Project (RAMIP). The SFLEs provide a decomposition for all pre-industrial through present day forcings into greenhouse gas-only forcing and anthropogenetic aerosol-only forcing, while one member additionally provides biomass burning (“black carbon”) aerosol-only forcing. The RAMIP, and a new black carbon-only forcing ensemble generated by this work, looks at how uncertainties associated with total aerosol emission, regional emission, and individual aerosol species impact near-term climate change. The project also develops novel aerosol-aware event attribution methodologies to better represent the impacts of aerosols on multivariate climate hazards both historically and in the future. To meet a growing need for a workforce trained in the computational and scientific theories specific to climate at the undergraduate level, this project integrates an undergraduate immersive, earth system modeling “field trip” framework and experiential earth system modeling curriculum for climate system science undergraduates with the goal of building the quantitatively climate-literate workforce needed to allow cross-sectoral operational climate planning. Finally, the project will mentor a doctoral student in aerosol processes and their role in multivariate climate extremes, climate hazard attribution methodologies, and undergraduate educational curriculum development. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

This Faculty Early Career Development (CAREER) award will fund research that attempts to enable transformative locomotion in insect-scale robots by endowing them with shape-shifting capabilities for achieving agile and adaptable locomotion in real-world environments. Natural terrains are complex and feature various types of clutters and confinements that restrict movement. In order to locomote through them effectively, robots must be small, as well as malleable and reconfigurable into terrain-specific body shapes just like their animal counterparts who expertly vary their body shapes and limb configurations. Thus inspired, this CAREER project aims to create a novel class of insect-scale exoskeletal robots that can passively conform to environmental restrictions, sense these constraints, and actively vary shape to be proficient at locomotion. In the long term, this research will improve the agility and adaptability of robotic systems and hasten their integration into real-world operations. Such shape-shifting miniature robots would squeeze through a pile of debris to autonomously locate survivors during search-and-rescue efforts, crawl inside a human body cavity to carefully remove tumors while assisting surgical procedures, or scurry between the blades of a jet engine to rapidly inspect and repair blade damage from bird collisions for the significant benefit of our society, health, and economy. The research is complemented by educational and outreach activities including project-based learning modules, lesson plans and training programs for K12 teachers, summer research programs for high school students, and interactive exhibits at distinguished science museums across Colorado. Research completed in association with this project intends create a novel class of shape-shifting insect-scale legged robots by focusing on three components – (1) Design body articulation to enhance adaptability via shape morphing and derive an empirical confined space locomotion model; (2) Incorporation of distributed sensing and actuation to actively control shape and improve agility; and (3) Development of a geometric mechanics modeling framework to find and optimize shape. By incorporating these innovations, this project intends to demonstrate novel insect-scale shape shifting robots capable of rapid, omnidirectional locomotion through laterally confined terrains, a first for legged systems. By leveraging insights from experimentally driven enhanced shape analysis and first principles-based effective shape finding, the project strives to take the first step towards rationally designing high-performance shape-shifting robots for applications beyond confined space locomotion. The project intends to demonstrate mapping and locomotion in GPS-denied confined environments utilizing a bioinspired tactile probe, a much-needed capability for effective complex real-world navigation. Finally, the project intends to contribute to several advances in integrated fabrication technology that enable new distributed sensing, actuation, and control in laminate designs at insect-scale for the first time to serve as a steppingstone towards realizing autonomic robotic structures embodying animal-like functionalities. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

This CAREER award funds four projects on agricultural markets. Each examines the effects of market frictions on whether or not farmers are able to sell in high value-added, export-oriented markets. The frictions include limited competition, lack of information, and financial market imperfections. Economic theory predicts that certain specific policies may reduce the impact of these market frictions and improve economic outcomes. Each research project tests whether or not these predictions are supported by real-world evidence in a specific agricultural market. Because farmers earn higher incomes when they are successful in producing these high-value products, the broader impacts of this award could include economic growth in regions dependent on agriculture as a main economic sector. The award also funds an education plan to mentor students and support international scientific collaboration. This CAREER award funds a research plan including four field experiments to study how market frictions at each step of agricultural supply chains impede structural economic transformation. Project 1 uses a randomized control trial (RCT) design to study the limited pass-through of quality incentives from world markets to upstream coffee producers to identify whether intermediary market power constrain quality upgrading by farmers. Projects 2 and 3 explore the role of search costs and information constraints in preventing agricultural small and medium enterprises (SMEs) from connecting with exporters and multinational buyers. Project 4 uses an RCT of an export loan facility to test whether agricultural exporters are constrained by limited availability of financing. In all projects, original data collection and administrative data access allow measurement of causal effects on directly impacted firms, on competitors, and on upstream suppliers. The results of this research project will help to identify ways to improve agricultural markets, increase agricultural incomes and accelerate structural transformation. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

Current machine learning methods are not capable of fully analyzing and interacting with satellite remote sensing data, limiting their ability to benefit society through real-world applications. Data collected by Earth-observing satellites differ significantly from image data (e.g., photos taken with a smartphone or digital camera) or text data (e.g., online articles or social media posts). Satellite data captures the Earth’s diverse and dynamic ecosystems, environments, and human activities that are constantly changing. These patterns and changes can be subtle or obvious, large or small, difficult or easy to see with the human eye. Satellites record these patterns in many different wavelengths and sensor types, which hold much more information than the visible color wavelengths humans see. This project will advance fundamental machine learning research methods for analyzing satellite data, unlocking its untapped potential for solving societal challenges including agriculture, conservation, and natural hazards. The project will develop new technologies that improve the performance and accessibility of satellite machine learning models for different applications, thus advancing scientific progress, human and environmental sustainability, and societal welfare. This award will develop (1) a hypermodal geospatial foundation model that accommodates diverse sensor modalities and input formats, (2) a novel algorithm for zero-shot mapping using natural language prompts instead of traditional fine-tuning, (3) a testbed to evaluate model robustness under realistic distribution shifts, and (4) a zero-shot evaluation algorithm that eliminates the need for ground-truth labels. These advancements will minimize the reliance on expensive data labeling and enable flexible and efficient interaction with machine learning models. The research will be iteratively validated through real-world deployments via NASA Harvest, NASA Acres, and other user-facing organizations implementing satellite solutions for global challenges. By treating ML research and deployment as a unified approach instead of siloed steps, this project pushes model evaluation into new regimes not typically explored in machine learning research. Broader adoption of this perspective in ML research will increase end-users’ trust and adoption of machine learning research. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

This award will support research in the field of algebra and representation theory with connections to theoretical physics. Representation theory is the branch of mathematics concerned with the study of symmetry using techniques from linear algebra, such as vectors and matrices. Classical examples of symmetry include rotations of a circle and reflections of a square. Modern research in mathematics naturally encounters more abstract notions of symmetry and requires the development of more advanced techniques for their study. The PI will identify and study concrete instances of these abstract symmetries and apply the results to advance the mathematical understanding of quantum field theory in dimensions two and three. The research will also develop new connections between algebra, representation theory, and low dimensional topology. This project integrates education, research, and career training opportunities for high school, undergraduate, and graduate students. Examples include creating a series of videos which will improve the algebra education of Utah's high school mathematics teachers-in-training, and organizing the graduate student focused Moab Topology Conference. In more detail, the PI will engage in three related research projects. Each project is motivated by and has direct implications for the long-standing conjecture that Rozansky-Witten theory, a three dimensional supersymmetric quantum field theory, defines a non-semisimple topological field theory. In the first project the PI will further develop and extend the theory of relative modular categories and their associated non-semisimple topological field theories to find new connections with homological blocks of 3-manifolds. In the second project the PI will develop the representation theory of affine L-infinity algebras and unrolled quantum groups and use the results to construct abelian approximations of Rozansky-Witten theory. In the third project the PI will extend his work on twisted equivariant matrix factorizations to construct non-semisimple, possibly unoriented, topological field theories for Landau-Ginzburg theory and explore their interpretation as Rozansky-Witten defects. The physical origins of the representation theories considered suggest many non-trivial interrelations. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

Most computers generate random-looking numbers without using any randomness at all; instead, the numbers are generated by a mathematical sequence that appears random. This might not be random enough for every application. Pseudorandomness, a key idea in theoretical computer science, involves constructing mathematical objects that exhibit random-like behavior while being created in a predictable way. These objects are essential for solving problems where true randomness is computationally costly or impractical to generate. The project addresses challenges such as developing efficient algorithms to determine whether complex mathematical expressions simplify to zero, a critical question in areas like algorithm design and circuit complexity. It also explores connections to error-correcting codes, which ensure reliable communication and data storage by correcting transmission errors. By advancing knowledge in these areas, the project contributes to foundational methods in computation and communication systems while supporting the national interest through fundamental research. The project also includes educational initiatives to create resources and courses that inspire the next generation of scholars in mathematics and computer science. The work spans computational complexity and coding theory, unified under the theme of algebraic pseudorandomness. In complexity theory, the project focuses on deterministic methods for polynomial identity testing in special classes of arithmetic circuits, addressing a central problem in derandomization. It also applies algebraic geometry to study structural properties of algebraic varieties that arise in computational problems, offering new perspectives in complexity theory. On the coding theory side, the research examines the combinatorial and algebraic properties of error-correcting codes, particularly list-decodable and list-recoverable codes, with the goal of developing more efficient constructions and improving their theoretical guarantees. By tackling these challenges, the project advances fundamental questions in pseudorandomness and contributes novel insights to the fields of complexity theory and coding theory. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

A group is a mathematical abstraction of symmetries of a physical object or a theoretical space. Groups are fundamental objects in mathematics that also emerge in various applications such as in computer science and physics. The algebraic notion of a group associates to a set a binary operation, like multiplication, which satisfies a list of axioms. Groups emerge naturally as symmetries of various types of concrete or abstract spaces in mathematics. There is an intricate relationship between the geometric properties of these spaces and the algebraic properties of their groups of symmetries. The PI will continue his investigation of the landscape of infinite groups that emerge as symmetries of the most natural spaces in mathematics, the circle and the real line. The PI will organize two research workshops aimed at graduate students, and two research experiences programs for undergraduates. These shall be aimed at training a diverse body of students to become future leaders in mathematics. These activities will incorporate computational methods into the students' mathematical exploration of the landscape of infinite groups. This project is jointly funded by Topology and the Established Program to Stimulate Competitive Research (EPSCoR). The PI will investigate the relationship between the algebraic structure of left orderable groups and the topological and dynamical properties of their actions on 1-manifolds and the cantor space. One goal is to investigate the class of finitely presented, infinite, simple groups, and exhibit new conceptual phenomena. This involves investigating notions such as uniform simplicity, and whether there is a finitely presented infinite simple group that acts on the real line by homeomorphisms. Finally, the PI will investigate a family of closely interconnected open problems emerging in combinatorial group theory. This includes a systematic study of normal generation in the class of finitely generated perfect groups, the conjectured existence of non-abelian free subgroups in non-indicable finitely generated left orderable groups, and fundamental groups of subcomplexes of aspherical 2-dimensional CW complexes. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

Estimating Power, Performance, and Area (PPA) earlier in the electronic design automation (EDA) flow would improve the Quality of Results (QoR) and reliability in chip design. The classical analytical or heuristic methods can be challenging to fine-tune, especially for complex problems. Machine learning (ML) methods have proven to be effective in addressing these problems. Graph Neural Networks (GNNs) have gained popularity since they are among the most natural ways to represent the fundamental objects in the EDA flow. However, with increased design complexity and chip capacity, an increasing performance gap exists between the extremely large graphs in EDA and the insufficient support from general-purpose hardware, such as mainstream graphics processing units (GPUs). This project aims to expedite the large graph machine learning on various EDA tasks, through a full-fledged development of efficient and scalable computing paradigms. This project's novelties are EDA domain knowledge-aware graph machine learning, training acceleration, and algorithm-hardware co-design and optimization. The project's broader significance and importance include: (1) to advance the field of machine learning in chip design, highlighted in National Artificial Intelligence Initiative; (2) to deepen the understanding of interactions among EDA domain knowledge, graph learning, and GPU acceleration; (3) to enrich the computer engineering curriculum and promote participation from undergraduates, underrepresented groups, and K-12 students in STEM fields through relevant programs. The project will develop a design paradigm for efficient, scalable and practical algorithm-hardware co-optimized solutions to significantly accelerate large graph machine learning on EDA tasks using a single GPU. This project consists of three coherent research thrusts: (1) to develop an algorithm-hardware co-optimized paradigm, focusing on restudying EDA graph features, introducing partitioning and selective re-growth methods, and tailoring GPU kernels for unified graph machine learning on EDA tasks using a single GPU; (2) to speed up single GPU for large circuit Graph Neural Network (GNN) training by implementing a tiled reversible architecture for low-memory training, and designing a maxK nonlinearity function to reduce computation costs; (3) to jointly integrate EDA domain knowledge, graph learning, and hardware optimizations to co-search for the appropriate hardware primitives and GNN compression strategies, as well as closely leverage the unique properties of circuit graphs. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

Fine-grained visual categorization involves identifying subtle differences between highly similar visual categories, such as distinguishing between two closely related bird species or recognizing different models of a car. While these capabilities are critical in a variety of fields, including biodiversity research, forensic investigations, and e-commerce, the task is challenging because differences between categories can be small, while variations within a category can be large. For example, two different bird species may look very similar, while male and female birds of the same species may look very different. Visual categorization in fine-grained domains is often treated as an image retrieval problem, where the label for a query image is determined based on the labels of the most visually similar images. An image retrieval approach typically performs better than standard classification approaches on fine-grained visual categorization tasks, especially in domains with very large numbers of classes. However, image-retrieval approaches often fail because a retrieved image that is visually similar is not necessarily from the same class, and potential images from the same class may not be retrieved due to low visual similarity. Moreover, the features learned by standard image retrieval models are often biased towards overall visual similarity rather than task-specific or domain-specific notions of importance. This limitation can hinder analysts and domain experts who may want to prioritize specific visual features due to their own expertise or intuitions. This project aims to improve image retrieval systems for fine-grained domains by aligning them more closely with how human experts perceive and prioritize differences, and enabling users to focus on the features most important to their task. While these innovations will be evaluated across a variety of fine-grained domains, they will studied through their integration into an existing real-world image retrieval system developed by the investigator that is used by analysts at the National Center for Missing and Exploited Children to recognize the hotels where victims of child sexual abuse and human trafficking are photographed. This project will address the limitations of traditional image retrieval in fine-grained domains by developing systems that better align with human notions of visual similarity while empowering users to dynamically guide retrieval processes. Ensembles of specialized models, each focused on a single visual notion, offer improved alignment with human judgments but are computationally impractical for real-time use, particularly in resource-constrained settings. To address this, the project will explore the use of knowledge distillation to integrate the varied knowledge of the ensemble models into a single model, while preserving flexibility for users to prioritize specific visual notions learned by individual models in the ensemble at query time. The project will also develop additional mechanisms for a user to dynamically refine search results. This line of work will have two directions: one where users specify their refinement by identifying visual features to prioritize, and one where the refinement is expressed in natural language. The project will also investigate the utility of subspace projection techniques as a pre-processing step for building task-specific indices from the embedding space of pre-trained vision language models. These innovations will enable both image-based and text-based retrieval systems where users can articulate preferences through natural language or visual cues, creating intuitive and scalable tools for specific fine-grained domains. Complementing these technical advancements, the project’s educational and outreach initiatives will focus on integrating machine learning competitions into undergraduate and graduate curricula to foster hands-on learning and critical thinking about real-world applications. Workshops will be organized to share best practices for using machine learning competitions as an educational tool, engaging educators, researchers, and students from a variety of backgrounds. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

Shellfishes use shells to protect themselves from predators and obtain nutrients from the water. Shells are tough since the compounds in them are harder than rocks. However, the toughness of shells can be easily affected by environmental conditions during the shell formation. Shells can become fragile when the shell formation occursin adverse conditions, such as high carbon dioxide (CO2) level or low salinity in water. With the effects on shells, the production of some shellfish species with high market values, such as the oysters, will be possibly declined in the coastal areas. The research goal of this CAREER proposal is to understand how the CO2 and salinity of water modify the Eastern oyster shell formation from a genetic level so that a method to help the oysters overcome the environmental effects during the shell formation. The result of this study will be applied to oyster aquaculture in the Gulf of Mexico. The project will also help with restoration of shellfish habitats on the coast. In addition, the project will involve the training of students from historically underrepresented groups in science at Texas A&M University – Corpus Christi via the educational goals of the project. The project will also provides research opportunities to community college students, as well as research opportunities for middle/high school science teachers in the biological sciences. It is well known that the shell development during the early life stages of bivalves is vulnerable to environmental stressors, but there is little understanding of the genetic responses of bivalves to environmental changes during shell development. The research goal of this CAREER project is to determine the molecular mechanisms by which bivalve shell formation is altered under ocean acidification (OA) and salinity fluctuation. The proposed approach is to 1) characterize changes in oyster shells under the stress of OA and salinity fluctuation by measuring the shell morphological changes, analyzing changes in shell composition, and identifying changes in matrix protein production of the shells; 2) identify the signaling pathway for shell formation response to OA and salinity stress by conducting transcriptomic analysis and calcium imaging with the primary cell cultures from the mantle tissue of the Eastern oyster; and 3) enhance the tolerance of the Eastern oyster shell formation to the environmental impacts by creating mutagenesis and transgenic strains. The education goals of the proposed project are to increase diversity in higher education and biological research, focusing particularly on encouraging Hispanic/Latino students to pursue advanced degrees in STEM fields. The approach will be to 1) create an innovative class for undergraduate and graduate students that involves multiple off-campus experts and facilities in teaching and student-learning outcome assessment; 2) providing research opportunities to undergraduate students and community college students; and 3) organizing a summer biological science section for middle/high school students and a training section for elementary/middle school teacher education. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

This CAREER project supports an integrated research and education approach to address the fundamental and applied challenges in solid-state nanopore-based single molecule counting platform towards the fully integrated point-of-care nucleic acid testing. This project is motivated by a growing demand for decentralized nucleic acid testing for applications ranging from infectious disease, food safety to homeland security. In recognition of using nanopore as simple as a single molecule counter and the fact that target nucleic acids can be sensitively and specifically replicated in numbers during the amplification reaction, this project seeks to develop a solid-state nanopore-based point of care nucleic acid testing device in which sample preparation is fully integrated. The proposed research activities will be synergistically integrated with educational activities to provide hands-on based education to inspire and train future STEM leaders and to increase public awareness of biosensing devices and their societal impacts. The overall research objective of this proposal is to explore a highly sensitive nanopore digital counting paradigm for point-of-care nucleic acid testing. Due to its potential for minimization and integration, solid-state nanopore sensing is a rapidly evolving field. Considerable effort has been made for developing various applications. However, translating solid-state nanopore sensors to practical settings has seen limited progress as compared to their biological counterparts adopted in the DNA sequencing, mainly due to the challenges in reproducible size control, introducing specificity, prolonged sensing time at low analyte concentrations, and the lack of integrated sample preparation. The proposed work aims to address these issues and explore a fully integrated solid-state nanopore digital counting paradigm towards nucleic acid testing at the point of need. The research activity will address four research aims regarding the sensing kinetics, nanopore fabrication, cartridge and system integration, and validation. This research is multidisciplinary in nature and spans the range from fundamental to applied research, with low- and high-risk components. Together, they constitute a complete research program to advance and translating the solid-state nanopore sensors. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

Employers across the United States are facing a crisis of labor supply, driven in part by changes in the working conditions employees desire and the wages they perceive as fair. Simultaneously, prospective homebuyers are faced with rising costs and fierce competition for housing – with aging adults, veterans, minorities, and low-income families hit hardest by rising home prices. This project seeks to understand how people assign value in these types of transactions (wages, homebuying), focusing on how individuals respond to competition and market changes. The project team is partnering with financial education programs on money management and first-time homebuyers to understand how differences between people – such as their tolerance for risks or willingness to delay purchases until a future date – predict their success on job and housing markets. The findings will allow us to better understand the psychology of value, create educational resources to help people adopt better strategies for finding jobs or housing, and identify public policy interventions that can alleviate housing and labor shortages. Across a series of experiments, this project tests computational models of pricing in multi-alternative, multi-attribute settings. These experiments manipulate simulated market competitiveness, delays, risks, and attributes of the choice alternatives in order to examine the dynamics of price-setting among individuals and test the cognitive mechanisms of pricing models. To improve the accessibility and efficiency of pricing model fitting and comparison, the project embeds them in neural networks trained to map observed pricing data onto generative model parameters. These will be added to online tools, teaching materials, and workshops to make them widely available. The pricing models will be used to estimate risk and delay aversion, bias, and pricing dynamics in multi-alternative choice experiments. The model parameters will also be used to predict the outcomes of job and home searches among participants in financial education programs. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

This project seeks to forge new connections between the theory of distribution learning and the study of diffusion models, one of the primary arms of the recent revolution in generative artificial intelligence (AI). We seek to develop a rigorous mathematical framework that can shed light on the remarkable effectiveness of these models in capturing complex, high-dimensional distributions. Our work will advance the frontiers of machine learning theory by enriching the field with new definitions for what these models are accomplishing and new algorithmic goals more closely aligned with how they are used in practice. We aim to provide a theoretical foundation that can guide the development of more efficient and controllable generative models and reshape our understanding of what makes learning tractable in high dimensions. As this effort inherently straddles research divides across theory and practice, we will complement our mathematical insights with extensive experiments on real-world data and with the development of new educational materials and research opportunities at the undergraduate and graduate levels to expand participation in this rapidly growing field. This program is built around three thrusts. The first is to develop novel analytic tools to understand what makes score estimation, the key subroutine on which diffusion models are built, possible for real-world data distributions. We will explore how to leverage properties like low-dimensional manifold structure and stability under perturbation to prove end-to-end algorithmic guarantees for score estimation. Our second thrust introduces a new paradigm of learning: instead of trying to generate samples that are statistically close to the ground truth, we aim to develop learners that can generate samples which are computationally indistinguishable. We will explore whether diffusion models based on "computationally optimal score estimation" can achieve this goal. This new framework promises to better align theoretical guarantees with practice, potentially explaining the empirical success of diffusion models on distributions that were previously considered intractable to learn. Finally, our third thrust will develop a unified theory for solving downstream tasks via foundation models, like sampling acceleration, model distillation, and guided generation. The algorithms we develop here will suggest new design principles for practitioners that could lead to faster and more versatile generative modeling. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

Bioelectronic sensors and therapeutic actuators enable the real-time monitoring and treatment of wound healing, diabetes, and other critical conditions; however, therapeutic functionality combining these sensors and actuators together remains limited to wearable devices, such as the insulin pump. Unlocking communication with implantable devices could revolutionize treatment for neurodegeneration, sepsis, solid tumors, and other morbidities, but it is hindered by a lack of clinically relevant communication protocols. These protocols either rely on power-hungry systems (e.g. Bluetooth) which demand large batteries and short lifetimes that inhibit implantation, or large antennas for wireless power and data transfer that are too bulky for patient acceptance. This project aims to develop and optimize a communication platform for multiple wearable and implantable devices to network with each other throughout all tissue layers in the body. The scientific goal of this project is to optimize an emitter and receiver network between electromechanical wearable sensors and implantable actuators via computational and experimental approaches, including in vivo experiments in rodent models with potential applications in sciatic nerve stimulation and controlled drug delivery. A second goal of this proposal is to promote bioengineering and STEM post-secondary education by creating high school physics lesson plans describing how bioelectronics relate to physics principles and by hosting high school students in the PI’s lab. This project could have profound impacts on public heath while also improving the toolsets available for fundamental biological analysis by simultaneously coordinating sensors and therapeutics throughout the body. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

Telomeres are important structures at the ends of chromosomes that protect DNA from degradation and in humans contribute to preventing cancer and premature aging. The length of telomeric DNA in different animals and plants varies substantially, but the significance of this variation for organismal integrity is not fully understood. This study will analyze the importance of having long or short telomeres for plants grown in good or stressful environmental conditions. Specifically, it will evaluate the number of seeds produced and the timing of flowering initiation as a measure of plant adaptation to stress. The project will also help extend hands-on research options for undergraduate students. It will also provide many opportunities for STEM-themed outreach activities to increase scientific literacy and to advance career preparation for the next generations of biology students. Telomeres are evolutionarily conserved protein-DNA complexes that cap linear eukaryotic chromosomes and ensure proper genome maintenance and stability. In humans, the length of telomeric DNA is often viewed as one of the most accurate cellular markers of biological age. However, in plants and many other organisms the significance of telomere length changes over organismal lifespan or in response to environmental stressors is largely unclear. Recent findings in the model plant Arabidopsis thaliana demonstrated that natural variation in telomere length influences reproductive fitness and the timing of floral transition. Nevertheless, the exact molecular and genetic mechanisms linking telomere length with plant developmental traits are unknown. Through several transcriptome and candidate gene analysis projects this study will elucidate the functional crosstalk between telomere length regulation and flowering time control. It will also characterize the effects of telomere length variation on plant fitness and decipher molecular mechanisms acting on extreme telomere length phenotypes. Finally, the project will aid in developing career-shaping research and educational experiences for undergraduate students. Overall, the proposed studies will help discover the intriguing functional roles that telomeres play in mediating life-history trait variation and trade-offs. This project is jointly funded by the Genetic Mechanisms Program, the Division of Molecular and Cellular Biosciences, and the Established Program to Stimulate Competitive Research (EPSCoR). This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

Glaucoma is one of the leading causes of irreversible blindness worldwide. This disease can progress unnoticed until vision loss occurs. This CAREER project aims to create innovative tools to unlock how glaucoma affects the nerve cells in the eye and to explore ways to detect and treat the disease earlier. The research includes developing a model that mimics glaucoma in animal eyes, using non-invasive technology to monitor nerve health, and applying advanced computer programs to analyze the data. These efforts will help uncover the cause of glaucoma and identify early signs of the disease. The educational aspect of this project will promote eye health awareness through hands-on workshops for K-12 students and new online resources, such as videos that teach at-home eye exercises. The project also includes developing teaching modules for college students to foster a learning environment accessible to all students. Through partnerships with local schools, industry, and community programs, this work aims to inspire future scientists, engineers, and healthcare professionals while contributing to new solutions for glaucoma care. This CAREER project aims to address key gaps in understanding glaucoma-induced neurodegeneration by developing an integrated platform combining bioengineering tools and neurobiological studies. The platform will include (1) an injectable microbead-based in vivo flow occlusion model to replicate glaucoma-like conditions and achieve sustained intraocular pressure (IOP) elevation, (2) a non-invasive electroretinogram (ERG) recording system to monitor retinal neural activity longitudinally, and (3) artificial intelligence (AI)-assisted algorithms for unbiased analysis of electrophysiological data. These tools will be applied to investigate the mechanisms driving retinal ganglion cell (RGC) degeneration and disruptions in lateral inhibition within the retina’s vision signal transduction pathway. The project will focus on optimizing biointerfaces in the flow occlusion model for reliable IOP elevation, advancing ERG systems for continuous monitoring of neural changes, and using AI to identify early-stage glaucoma biomarkers. Additionally, pharmacological and genetic regulation of specific inhibitory neuron populations will be combined with ERG and IOP monitoring to study how hyperactivation of inhibitory neurons contributes to RGC loss. The resulting platform will provide insights into the cellular and molecular mechanisms of glaucoma, enabling earlier detection and identifying potential therapeutic targets. This integrated approach aims to improve outcomes for glaucoma patients and advance technologies for vision science. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

The survival of coral reef ecosystems depends on their capacity to adapt to rapidly warming oceans. This project aims to understand and predict these critical adaptation processes by developing predictive tools for coral reef conservation while creating educational pathways in marine science. Through workshops, internships, and public education programs at the Hawaiʻi Institute of Marine Biology, this research engages students and community members in understanding how corals may adapt to changing environments. These educational initiatives focus on student mentorship and training, while research outcomes directly inform conservation strategies to protect reef ecosystems for future generations. This research aims to develop a mechanistic theoretical framework that models coral adaptation by integrating three key mechanisms: larval connectivity between reefs, genetic adaptation of coral hosts, and dynamics of their symbiotic algae. The project will construct a series of mathematical models spanning multiple scales, from within-host symbiont communities to reef-network connectivity patterns. These models will quantify how different coral and symbiont characteristics influence adaptation patterns across reef networks under various climate scenarios. The framework can be applied to Hawaiian coral reefs using local oceanographic data to assess their adaptive potential through 2100. Model predictions aim to identify vulnerable coral species and reef areas, as well as potential refugia. By revealing the key mechanisms driving coral adaptation across scales, this work will provide critical insights for coral reef conservation under rapid environmental change. This project is supported by the Biological Oceanography Program in the Division of Ocean Sciences and the Mathematical Biology Program in the Division of Mathematical Sciences. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

This program uses gravitationally-lensed quasars to study the nature of dark matter. By using high spatial-resolution spectroscopy, Nierenberg will measure positions and flux ratios for 200 multiply-lensed systems -- an order of magnitude increase in the existing number of similar measurements. These data will be used to determine the relative number of dark matter halos well below the mass threshold for galaxy formation, the presence of dark matter particles with high enough velocities to impact galaxy formation and galaxy clustering, and to confirm or exclude the existence of entirely dark halos. Research-integrated educational components include training and mentoring of a postdoctoral scholar and graduate students, a summer academy in STEM topics targeting 4th and 5th graders, course curriculum development in General Relativity, and public lectures series on Nierenberg’s research. Integral field spectroscopy with adaptive optics on multiple large-telescope facilities will be used by Nierenberg to constrain the distribution of dark-matter halos from the narrow emission-line flux ratios of quadruply lensed quasars, selected from Euclid and LSST imaging. Lens flux ratios are sensitive to the second derivative of the gravitational potential, offering unique sensitivity to the number and distribution of low-mass halos. A primary result from this work will be a definitive measurement of a possible turnover mass (at 106 solar masses) in the halo mass function, indicative of warm dark matter. The analysis approach is statistical, using forward modeling and Bayesian computing methods to evaluate the highly stochastic mapping of dark matter distribution parameters to predicted flux ratios. The educational components of this program are intertwined with the research themes, providing instruction and mentoring in the scientific method, scientific computing and dark matter research to students spanning elementary school through postdoctoral career phases. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

Markov chains are an important family of stochastic processes used in many scientific fields, including sampling and optimization algorithms and physical modeling of thermodynamics. In recent decades, their mathematical study has been oriented around the time to converge to an equilibrium state from a worst-case initialization, known as the mixing time. Often, however, the worst-case mixing time is much larger than the convergence time from typical initializations. This project will develop the theory of convergence times from commonly encountered initializations when worst-case mixing times are slow. The research program will be complemented by educational activities, including a summer school for graduate students focused on the interplay between high-dimensional probability, data science, and machine learning. The project will also support the PI’s engagement with the broader scientific community around developing solid theoretical foundations to parallel the rapid acceleration of artificial intelligence capabilities. The project also provides research opportunities for graduate students. The research aims of the project are focused on developing mathematical tools for studying equilibration of Markov chains in high-dimensional spaces from nice initializations (e.g., random or high-temperature) when worst-case mixing times are exponentially slow in the dimension. Markov chains to be studied include families appearing in (1) Markov chain Monte Carlo sampling in computer science, (2) low-temperature Glauber dynamics in statistical physics, and (3) optimization algorithms on non-convex loss landscapes in machine learning. A core goal is to put the scientific understanding of simulated annealing, metastability, and the role of initialization in high dimensions on a more rigorous mathematical footing. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.