San Diego State University Foundation

NSF Awards · FY 2026 · 2026-09

As Research and Engineering (R&E) networks have evolved to support an array of data-driven science, they have become highly converged and inter-dependent with the wider Internet.  This convergence has resulted in remarkable efficiencies, data and resource sharing, and distributed collaborative research, but has also imparted network complexity and dependencies.  Because R&E networks carry critical science traffic, provide critical services, and enable production workflows, hidden dependencies can result in correlated failures and security vulnerabilities.  The “Cyberinfrastructure Robustness Understanding through data eXtraction” (CRUX) project takes a data-driven approach to better understand the robustness of R&E networks. CRUX performs continuous, network measurement-based data collection and curation to enable studies of the robustness of modern converged R&E networks. The two primary thrusts of CRUX are: 1) innovative active and passive measurement techniques to map the R&E ecosystem and gather currently unavailable data; and 2) dissemination and active engagement to help the community utilize the datasets.  Our collection methodology includes new techniques for discovering and cataloging R&E organizational resources, state-of-the-art active measurement techniques for rapid and rich connectivity mapping, passive measurements for attack and misconfiguration detection and characterization, and diverse meta-data. These disparate measurements are fused and augmented into three curated datasets – D1) R&E Ecosystem; D2) R&E Net Connectivity; and D3) R&E Net Activity – that provide the ability to draw new insights into the composition, robustness, and cybersecurity activity of R&E networks. Such insights will be valuable for operators, policy makers, and network and security researchers. 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 Awards · FY 2026 · 2026-06

Soil-associated organisms represent a large proportion of Earth’s biodiversity, but knowledge about these organisms is very limited, with most species yet to be discovered and studied. Many soil-associated organisms are found in only small areas. For example, some are found in areas no larger than 300 square miles, and even less is known about these types of organisms. Spiders are important components of this soil-associated biodiversity. This project seeks to discover and describe part of the vast diversity of locally distributed spiders inhabiting soil in California using extensive fieldwork and museum work, combining with modern tools in digital microscopy and genomics. Understanding biodiversity and species histories across multiple groups of soil spiders will provide researchers with valuable information for recognizing and prioritizing regions in California having high levels of unique biodiversity. This project will also conduct a course focused on spider identification and biotechnology, via a bioinformatics bootcamp to help train the next generation of students in biodiversity sciences and spider research using modern genomic tools and spatial data. Micro Short-Range Endemic taxa (microSREs) represent species with limited distribution areas, usually less than 1,000 square kilometers. Because of microhabitat specificity, microSRE are closely tied to landscapes, more so than most other terrestrial animal taxa. This close association with landscapes implies extreme dispersal limitation, making them extremely rare at the global level and of conservation relevance. MicroSREs present special challenges in revisionary systematics but also provide profound and unique insight in studies of the speciation process, regional biogeography, spatial phylogenetics, and landscape conservation. Because of ecological niche conservatism, microSRE taxa illustrate processes of a “divergence under niche conservatism” model of speciation. This project aims to conduct integrative revisionary systematics and address coincident evolutionary and biogeographical questions for parallel radiations of related cybaeid spider genera from the endangered California Floristic Province (CAFP). This project involves extensive fieldwork in the CAFP, emphasizing litter and soil surface sampling. Freshly collected samples will fill sampling gaps for revisionary systematics, provide fresh specimens for phylogenomics, and result in multiple 1000s of expertly curated “by-catch” arthropod specimens, to serve as important genomic and morphological vouchers in studies of other rare and endemic taxa. Multiple phylogenomic datasets of largely co-distributed taxa will be used to test riverine barrier and elevational tiering biogeographic hypotheses, and be used to conduct spatial phylogenetic analyses, helping to understand multi-taxon phylogenetic diversity and endemicity hotspots in the California Floristic Province. 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 Awards · FY 2026 · 2026-06

Living systems have evolved highly precise ways to sense and respond to microbial cues, yet the rules governing these interactions remain unclear. Understanding these mechanisms is essential not only for biology, but also for advancing biotechnology platforms that harness microbial products to control cell behavior. This project addresses a fundamental question in biology: how do animals interpret microbial cues to regulate development. By revealing how conserved biological pathways link bacterial sensing to developmental decisions, this work will advance understanding of host–microbe interactions. The results have the potential to inform strategies for marine ecosystem restoration, including enhancing larval recruitment in degraded habitats, while also establishing foundational principles for biotechnology applications that harness microbial cues to control biological processes. Discoveries from this research are thus expected to enable emerging biotechnology platforms, illustrating how fundamental insights can translate into new environmental and biomedical applications. The project will also support workforce development through hands-on research experiences for undergraduate and graduate students, including a course-based program that engages early-career students in discovery-driven science. In addition, this project advances NSF’s priorities in Biotechnology. This project uses the marine tubeworm Hydroides elegans as a model system to define how animals detect bacteria and initiate metamorphosis. The central hypothesis is that larvae use conserved pathways associated with the innate immune system, including components of the Toll-like receptor (TLR)/MyD88 signaling pathway, to distinguish between beneficial microbial cues that trigger development and pathogenic signals that activate defense, thereby directing distinct biological outcomes through NF-kB–dependent programs. To test this, the project will (1) identify and functionally characterize NF-kB homologs required for bacteria-induced metamorphosis, (2) determine whether beneficial and pathogenic bacteria activate overlapping or distinct host signaling pathways, and (3) identify the receptors responsible for detecting bacterial cues. The approach integrates RNA interference (RNAi), gene expression analysis, and bacterial genetic tools to causally link microbial signals to host developmental responses. By defining the molecular pathways that connect bacterial detection to developmental activation, this work will establish a mechanistic framework for how environmental microbes regulate animal life histories and reveal conserved principles of immune–developmental signaling. 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 Awards · FY 2026 · 2026-05

This project is motivated by the unprecedented complexity faced today in ensuring the dynamic security of electric power networks hosting heterogeneous electric energy resources (EERs). The project aims to develop two types of distributed intelligence (DI), namely DI-1 and DI-2, for assessing and enforcing system-level, electromagnetic transient (EMT) stability of a broad class of electric power networks in the presence of large disturbances. The project will bring transformative change in how to reliably operate electric power networks with both legacy and next-generation resources. This will be achieved by DI-1, a power-electronic interface that can enforce stability without reprogramming of EER controllers, and DI-2, a software module that self-checks the compatibility of an EER with its host network and adaptively tunes EER controllers to ensure this compatibility. The intellectual merits include: 1) Both types of DI will be able to address the high-order, nonlinear dynamics governing power networks with heterogeneous EERs. 2) DI-1 can self-check and self-enforce the compatibility of EERs with their host grids without reprogramming the controllers of EERs. 3) DI-2 can provide grid operators with real-time situational awareness of grids’ resilience to large external disturbances. 4) EMT-Cloud is the first-of-its-kind, open-access cloud platform enabling nationwide undergraduate and graduate researchers to remotely perform EMT simulations and hardware-in-the-loop (HIL) tests. The broader impacts include: 1) The proposed intelligence will enhance the U.S. grids’ resilience under large disturbances by providing system operators with situational awareness on grid stability and by preventing heterogeneous EERs from destabilizing the systems. 2) EMT-Cloud provides nationwide student researchers with an efficient platform for prototyping their power system research ideas requiring HIL/EMT simulations. 3) The project will introduce a course featuring Inverter-Based Resource (IBR) modeling, decentralized/distributed control, and real-time simulations. The project will also include two hands-on training workshops open to students nationwide. These efforts will contribute to workforce development in the U.S. This project includes three synergetic thrusts. Thrust 1 aims to design DI-1 for Non-Manufacturer Parties (NMPs). DI-1 checks and enforces local stability conditions without accessing or reprogramming EER controllers. Thrust 2 will focus on design DI-2 for EER manufacturers. DI-2 will be able to significantly accelerate transient stability assessment through an asynchronous learning-falsification method. DI-1 and DI-2 will support interoperability among legacy and next-generation resources, and they can stabilize their host networks collaboratively regardless of the availability of communication. With the two types of DI, Thrust 3 will establish a framework that enables grid operators to stabilize the networks with minimal changes to the network topology. 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 Awards · FY 2026 · 2026-04

Heat waves are increasing across the U.S., altering the health of coastal ecosystems and, by extension, the health and welfare of the American public who depend on coastal resources. The success of coastal systems may depend on species that create habitat, protecting the overall ecological community from heat-wave impacts. Seaweed canopies and shellfish beds can provide shade and retain moisture, maintaining a cooler, damper environment that buffers the effects of heat waves. This project evaluates the ability of two of the most abundant habitat-forming species on the U.S. coastline – rockweeds and mussels – to maximize the survival of coastal species experiencing increasing heat waves along the west coast of North America from California to Alaska. The project combines observations during natural heat-wave events, heating experiments, and physiological studies of habitat-formers and the plants and animals they support to identify the vulnerabilities and refuges within the coastal ecosystem during these extreme events. The project will also provide training opportunities for students, expand course-based research experiences, and support outreach and knowledge exchange with environmental management agencies and the public. Together, these activities strengthen scientific capacity and improve understanding of coastal ecosystem resilience in a changing environment. To evaluate the ability of coastal foundation species to mitigate impacts of extreme heat events, the investigators test the hypothesis that foundation species are more tolerant of heat waves than the associated species that they support. Rockweeds and mussels are abundant foundation species likely to withstand stressors that other, facilitated species are unable to tolerate. But intensification of heat-wave events could push foundation species beyond their limits. There is a critical need to understand the potential for foundation species to mitigate impacts of heat waves or for cascading local extinctions to occur associated with their losses. The investigators combine observations during natural heat-wave events and in situ heating experiments with measurements of environmental conditions and physiological performance of foundation species and associated species. The objectives are to (i) quantify foundation species’ facilitation of associated communities by mitigating environmental stress; (ii) determine susceptibility of foundation species to heat-wave conditions and how their loss reshapes intertidal communities; (iii) identify physiological mechanisms underlying susceptibility of foundation and associated species to temperature stress; and (iv) predict outcomes of intensifying heat waves for foundation species and associated communities. The results of this research lay a strong foundation for anticipating the impacts of extreme heat events on the health of coastal ecosystems. 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 Awards · FY 2025 · 2025-10

The Advanced Water Resilience through Collaborative Operational Strategies (AWRCOS) Incubator focuses on improving water management in California’s San Joaquin Valley, a region of major agricultural importance. Communities and farms in the area face recurring challenges related to drought, flooding, and groundwater declines. This project combines modeling, water systems engineering, machine learning, and forecasting tools to develop practical solutions in partnership with local agencies and stakeholders. Strategies under evaluation include Forecast-Informed Reservoir Operations and Managed Aquifer Recharge, which rely on real-time weather and streamflow data to guide water storage decisions, reduce the likelihood of flood damage, and support groundwater levels. The project emphasizes collaboration between researchers, local decision-makers, and community members to ensure that scientific tools are tailored to real-world needs. The approaches developed through this effort are intended to support both regional planning and broader resource management goals. By connecting forecast tools with operational decisions, AWRCOS aims to strengthen local capacity to respond to variable weather conditions and support continued agricultural productivity. The Advanced Water Resilience through Collaborative Operational Strategies (AWRCOS) Incubator addresses the increasing complexity of water management in California’s San Joaquin Valley, where weather and extreme stressors are compounding risks to agriculture, ecosystems, and communities. The project aims to develop adaptive water management strategies that respond to deep uncertainty in both weather and policy conditions. The central objective is to integrate Forecast-Informed Reservoir Operations and Managed Aquifer Recharge with Dynamic Adaptive Policy Pathways, enabling flexible decision-making that evolves over time in response to changing conditions. The project leverages significant prior investment in the region including Coupled Model Intercomparison Project 6 (CMIP6) data downscaled to 3 km developed for California, robust socio-economic data available for the region, and recent advancements in seasonal and subseasonal forecasts. Phase 1 prioritizes engaging regional stakeholders to gather detailed feedback and ensure that the solutions are developed to meet the needs of local water agencies, the agriculture industry and communities. Together these data and methods will be used to develop a decision support tool design and address the challenges and needs of the region that can be explored for the broader southwest. 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 Awards · FY 2025 · 2025-10

Provenance tracks the origin, creation, usage, and modifications of data, and is essential for verifying data integrity and conducting security audits. It supports compliance with strict privacy regulations in sectors like healthcare, finance, and government by maintaining detailed records of data activities. However, detailed provenance graphs can expose sensitive information and are vulnerable to adversarial attacks. Differential privacy provides a way to minimize privacy risks by reducing the impact of individual data contributions on analytical outcomes, offering robust protection against adversaries with extensive knowledge. However, applying differential privacy in practice remains challenging due to the need of balancing data utility with privacy. The project's novelty lies in bridging the gap between differential privacy's theoretical benefits and practical implementation in provenance graphs. The project's broader significance and importance include translating the findings into refined methods and synthetic datasets, and disseminating outcomes to sectors like healthcare, finance, and public safety through educational workshops and seminars. This project is structured around two main thrusts. Thrust 1 focuses on 1) identifying privacy risks in current provenance-based machine learning anomaly detectors by systematically applying property inference attacks, membership inference attacks, and graph reconstruction attacks; and 2) developing snapshot-level differential privacy with implications that align with enterprise privacy policies, compliance requirements, and performance needs . Thrust 2 aims to develop practical solutions by generating differentially private graphs, through the design of a subgraph synthesis method that addresses the dense correlations and large scale of raw provenance 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 Awards · FY 2025 · 2025-09

The construction industry is poised for a major digital transformation with technological innovations that are set to improve the safety, productivity, and sustainability of the field. To leverage these advanced technologies to their full potential, there is an urgent need for skilled professionals who understand both construction and digital technology. This National Science Foundation Research Traineeship (NRT) award to San Diego State University (SDSU) in collaboration with the University of Virginia (UVA) will develop a comprehensive graduate program in smart construction technologies. The project anticipates training of one hundred and thirty‐five (135) MS students, including thirty-one (31) funded trainees, from fields such as civil engineering, electrical engineering, mechanical engineering, and computer science. By bridging the gap between traditional construction practices and emerging digital innovations, the program will help create safer worksites, better-performing buildings, and more sustainable infrastructure while opening construction careers to more groups of professionals. The SCIBER-CT (pronounced “cyber city”) NRT program will integrate three technological focus areas within modern construction practices: artificial intelligence and machine learning, robotics and automation, and advanced communications and immersive technologies. The educational framework includes an 18-unit certificate program featuring specialized coursework, research components, and internships, with plans to develop a full master’s degree following a feasibility study. Students will participate in professional development modules covering essential skills like communication, team science, and ethics. Experiential learning through industry internships and hackathons will allow students to apply their technical knowledge to real-world construction challenges. These elements are designed to cultivate T-shaped professionals who not only excel in their areas of specialization but are also skilled in cross-disciplinary collaboration. The cross-institutional collaboration leverages SDSU expertise in advanced construction technology and UVA’s established experience with interdisciplinary graduate education in cyber-physical systems. This partnership creates a model for expanding advanced graduate education opportunities while addressing critical workforce needs. The promise of the SCIBER-CT program is to transform graduate training in construction by creating a scalable, industry-relevant educational ecosystem that advances both scientific knowledge and societal benefits. The NRT Institutional Partnership Pilot (NRT-IPP) is a collaborative effort between the Directorates for STEM Education (EDU) and Technology, Innovation, and Partnerships (TIP) to support research and education projects with high industry relevance in emerging fields. 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 Awards · FY 2025 · 2025-09

Currently, about half of the world’s population, about 4 billion people, live in areas with a risk of dengue infection, with further increased public health concerns due to recent evolutionary adaptation of dengue-transmitting mosquitoes to colder places. This project develops and uses mathematical models and computational methods, including the novel Math-model Informed Neural Networks (MINN) based on emerging mathematics in mosquito-dengue biology. Proper management guidelines, identified through data-driven models, help healthcare professionals mitigate the burdens of dengue infection, thereby improving the quality of life for dengue-infected patients and their families. The outcomes of this project not only fundamentally advance the fields of mathematical biology and quantitative biology but also have a simultaneous broad and highly positive societal impact. In addition, this project offers extensive interdisciplinary research training opportunities for undergraduate and graduate students in mathematics and biology. The project will expand research and educational opportunities to various programs for students, as well as junior and senior researchers, and will incorporate the research into an interdisciplinary mathematical biology course. This project will focus on three aims: (a) Develop Math-model Informed Neural Networks (MINN) capturing emerging mathematics in climate-dependent mosquito-dengue biology. (b) Analyze models and develop MINN-based methods to estimate epidemic thresholds. (c) Develop MINN-based user-friendly online platforms for public health policy evaluations and healthcare accessibility. The novel models, validated using data from our collaborators (biologists/environmentalists) from Nepal, will incorporate an experimentally observed mosquito life cycle and dengue transmission. The models and related MINNs will be used through a user-friendly online platform to evaluate public health policies and calculate healthcare accessibility for dengue control in spatially heterogeneous environmental conditions. This contribution will have a significant positive and practical impact on developing public health policies to prevent dengue virus infection, as well as advance the development of sophisticated mathematical and machine learning models to help explain the role of the environment and mobility in the complex biological systems of mosquito-dengue interactions. 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 Awards · FY 2025 · 2025-09

Eleven percent of the world's recorded earthquakes occur in Alaska, with many notable historical events including the 1964 M9.2 Alaska earthquake, the second largest earthquake ever recorded, as well as the 2018 M7.1 Anchorage earthquake. Earthquake hazards in south-central Alaska are a serious concern because the area includes the city of Anchorage where more than a third (~280,000 people) of the state’s population live. To facilitate estimating seismic hazard for the region from future large earthquakes, this project will assemble the first version of a multi-scale three-dimensional Community Velocity Model (CVM) of the subsurface of south-central Alaska. The CVM will be built with existing model components, calibrating its parameters by comparison of simulated and recorded waveforms for small local earthquakes. The CVM will be freely available to local agencies, scientists, and communities, bridging advanced modeling with practical applications such as assessing the seismic hazard to critical infrastructure in this region. The project will train a graduate student in seismic wave propagation simulation, fostering the next generation of geoscientists. The researchers plan to assemble and distribute the first version of a multi-scale Community Velocity Model (CVM) for the south-central Alaska Region, SCAR-CVM V1.0, from existing model features. The accuracy of such a CVM is fundamental to estimates of ground motions for seismic hazard analysis using physics-based wave propagation. Features in the proposed CVM will include 3-D tomographic velocity variation, high-resolution surface topography, the depth to basement in the Cook Inlet basin, generic basin velocities combined with near-surface constraints in the sediments, and a low-velocity weathering layer constrained by Vs30 values. Modeling of 0-1 Hz ground motions and comparison to seismic observations for six local M4.6-5.9 earthquakes will allow calibration of the velocity and attenuation structure in the model, as well as help assess the efficacy of the model for waveform prediction. This project will demonstrate to what extent the model can forecast geographically specific ground motion patterns such as extended durations and amplitudes, or amplification above the sedimentary basins of south-central Alaska. These are all capabilities needed for physics-based seismic hazard analysis. The work will also incorporate into the CVM the best available geometry of the subducting slab south-southeast of Anchorage, allowing future simulations of megathrust simulations such as the 1964 M9.2 earthquake, as well as intra-slab events such as the 2018 M7.1 Anchorage earthquake. The planned CVM will be the first iteration of any future improvements when new parameter constraints become available for studies such as nonlinear analysis of the soils and critical lifelines, as well as help enhance seismic risk preparedness, bridging advanced modeling with practical applications. 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 Awards · FY 2025 · 2025-09

This project addresses a major challenge in next-generation autonomous systems: enabling unmanned aerial vehicles (UAVs) to physically interact with moving objects in mid-air. This capability has the potential to unlock transformative applications such as mid-air refueling, maritime rescue, and dynamic target interception. These tasks are difficult due to constantly changing environments and require high levels of precision, adaptability, and reliability. By tackling this challenge, the project supports national priorities in aviation safety, disaster response, and transportation logistics. The research will also offer hands-on learning experiences for graduate and undergraduate STEM students. The technical focus of this project is on advancing cyber-physical systems through a unified framework for aerial manipulation in dynamic environments. The research will develop: (1) a physics-informed modeling framework that improves UAV flight dynamics prediction under real-world conditions, (2) a safety-assured motion planning approach for computing adaptive, collision-free trajectories in uncertain and dynamic scenarios, and (3) a novel, energy-absorbing manipulator that enables UAVs to securely grasp and interact with moving objects. These goals will be achieved through the integration of machine learning, robotic system design, and optimization methods. The resulting algorithms and hardware will be validated through experiments and are expected to significantly expand the capabilities of UAVs for real-world operations in logistics, emergency response, and airborne transportation. 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 Awards · FY 2025 · 2025-08

The failure of the right ventricle can indicate a higher risk of mortality in diseases like pulmonary hypertension and heart failure with preserved ejection fraction. People with these diseases often report exercise intolerance, which suggests that their response to physical stress is impaired. This research project aims to improve our understanding of the mechanical behavior of the right ventricle and how it affects its pumping function. The concept of viscoelasticity, which refers to the immediate (elastic) and delayed (viscous) resistance of cardiac tissue during contraction and relaxation, will be used to study how cellular behavior and the overall performance of the heart are affected. The research will include a comprehensive analysis of the viscoelasticity of the right ventricle tissue and its implications in organ function. The findings of this study could be used to the care and management of approximately 6.7 million heart failure patients, which are projected to increase with the aging population. The research team plans to engage young minds and advance the education of the next generation of bioengineers, with a focus on mentoring women and first-generation students to enhance diversity in the STEM workforce. The aim of the research is to discover how the ventricle wall viscoelastic properties change under acute stress and chronic pulmonary hypertension and impact organ function. To characterize the tissue biaxial viscoelasticity with disease progression and at varied heart rates, the passive viscoelasticity of the right ventricle will be measured experimentally by stress relaxation and cyclical tensile tests and simulated computationally with our validated models. To reveal the cellular and extracellular components’ contributions to tissue viscoelasticity, drug treatment will be performed to depolymerize microtubules and degrade collagen and changes in tissue viscoelasticity will be quantified. A new multiscale constitutive model will be developed to predict the viscoelasticity of myofibers and collagen as well as collagen recruitment. Finally, organ function changes will be measured with right ventricle viscoelasticity altered by chronic hypertension and acute stress. Educational activities include the distribution of various STEM kits to elementary and middle-to-high school students in northern Colorado, participation in existing K-12 public outreach events on campus. Additionally, there will be training and mentoring of underrepresented undergraduate and graduate students. 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 Awards · FY 2025 · 2025-08

An increasing umber of wildfires that approach has made the Wildland-Urban Interface (WUI) one of the most dangerous regions in the United States. When WUI fires burn through communities, the destruction extends far beyond structural losses. Unlike wildland fires that mostly burn trees and grass, WUI fires ignite plastics, electronics, and household furnishings, releasing a complex mix of harmful air pollutants. This project will identify the pollutants that are released and quantify the amount produced when materials typically present in WUI environments burn. Controlled burns will be conducted in the laboratory to understand how these pollutants move through homes, settle on surfaces, and become airborne again, potentially exposing returning residents and workers. Air samples from homes affected by the 2025 Los Angeles fires will also be collected. Results of the project will improve our understanding of indoor air quality after a fire and help guide safer clean-up and reoccupation strategies. Although the frequency and severity of WUI fires have increased dramatically in recent years, critical gaps remain in our understanding of the chemical emissions and exposure risks they create. Most existing studies focus on wildland vegetation fires and fail to account for the complex mixture of pollutants released when manmade materials burn. Additionally, there is limited knowledge of how combustion conditions influence pollutant formation in WUI fires, and even less is known about how toxic emissions infiltrate homes, persist on indoor surfaces, and re-emerge after cleanup. To address these challenges, this project has three integrated objectives: (1) characterize multi-phase organic and inorganic emissions (including speciated particulate metals, volatile organic compounds (VOCs), carbonaceous particles, as well as gases such as carbon dioxide (CO2) and nitrogen dioxide (NO2)) from materials commonly found in WUI environments under controlled combustion conditions that simulate real-world fire scenarios; (2) investigate the penetration of outdoor fire emissions into indoor environments, examining the roles of environmental and human factors, chemical signatures, spatial gradients, source apportionment, and long-term pollutant dynamics; and (3) develop and validate a computational model that simulates indoor pollutant dynamics, including dispersion, deposition, resuspension, and off-gassing. This research will advance fundamental understanding in chemical, atmospheric, and exposure sciences by bridging key knowledge gaps in the emission, transport, and indoor infiltration of pollutants from WUI fires. By integrating real-time field measurements using innovative instrumentation with bench-scale and larger-scale combustion experiments and advanced exposure modeling, the project will provide a comprehensive framework for understanding the fate of WUI fire-emitted pollutants across environments. Furthermore, this research will refine exposure modeling frameworks by incorporating empirical data on time-varying indoor concentrations and surface interactions, enabling the simulation of critical yet understudied human exposure processes in WUI fire-impacted communities. The outcomes will reshape our understanding of post-fire air quality hazards and set the foundation for predictive models of WUI fire emissions and their health impacts. 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 Awards · FY 2025 · 2025-08

Elastomeric polymers are widely used in various industries, ranging from biomedical devices to automotive safety systems, due to their flexibility, durability, and ability to absorb energy. However, current understanding holds that these materials respond primarily in a highly elastic manner and tend to fail in a brittle fashion under extreme conditions. Research supported by this project aims to transform that understanding by uncovering how elastomers behave when subjected to extremely high deformation rates, such as those experienced in ballistic or shock-loading environments. By doing so, this research looks to provide insight into the ability of elastomeric polymers to undergo permanent, plastic deformation, contrary to conventional belief. These findings could lead to the design of new polymer-based structures with improved performance in impact mitigation, flexible electronics, and rapid manufacturing. Additionally, the project will contribute to training students in cutting-edge experimental and computational techniques, integrating research into engineering education, and promoting an exchange program between the two research laboratories. This research investigates the mechanisms of plastic deformation and failure in elastomeric polymers under ultra-high strain-rate loading using a novel laser-based shock technique coupled with bulk spectroscopy that can strategically shape and tune the rate of deformation. The experimental approach enables repeated shock loading of the same location to simulate and study shock fatigue and damage accumulation. Through advanced characterization methods, including spectroscopy, microscopy, and thermal and mechanical testing, the project seeks to reveal how molecular networks evolve under extreme conditions. These observations look to inform the development of a new multiscale computational model that couples molecular-scale deformation, damage accumulation, and rate-dependent mechanical response. The model intends to incorporate thermodynamic principles and micromechanical behavior to predict how different classes of elastomeric polymers behave under shock loading. The combination of experimental discovery and theoretical modeling is expected to produce a fundamental shift in the understanding of polymer plasticity and damage evolution, with broad applications across manufacturing, defense, and biomedical fields. 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 Awards · FY 2025 · 2025-08

Artificial intelligence (AI) systems are increasingly used to understand and replicate human decision-making in urban environments, from ride-sharing and delivery to traffic planning. These systems rely on spatial-temporal imitation learning (STIL), where models learn from human mobility patterns to optimize services. However, as these models ingest massive amounts of sensitive user data (e.g., GPS traces), new challenges arise in ensuring privacy and compliance with emerging “right-to-be-forgotten” regulations. This project will develop new technologies that enable STIL models to selectively forget specific user data without retraining from scratch. The novelty of this project lies in its ability to effectively erase individual trajectories or behaviors embedded in complex user data, while preserving the utility and accuracy of the original models. The proposed technologies will empower users with control over their data and help ensure that AI systems used in urban intelligence are not only effective but also privacy and legally compliant. This project will pioneer a new research direction in machine unlearning within the spatial-temporal imitation learning (STIL) domain by introducing the SIFT (Spatial-temporal Imitation ForgeTting) framework. SIFT is designed to address three key challenges: (1) overlapping and dispersed trajectories that make precise data unlearning difficult, (2) spatial-temporal heterogeneity and sparsity that can degrade model performance when unlearning data from data-sparse regions, and (3) lack of theoretical guarantees for effective unlearning. The research will develop innovative methods including negation trajectory generation and reward shaping to remove privacy-sensitive data, spatial-temporal partitioning and similarity-based data augmentation to handle data sparsity and variability, and differential privacy techniques to ensure provable unlearning guarantees. These contributions will be evaluated using large-scale, real-world datasets from ride-hailing and transit systems. The project will also integrate unlearning research into undergraduate and graduate courses. 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 Awards · FY 2025 · 2025-07

Brain-computer interfaces (BCIs) enable communication between the brain and external devices for individuals with neurological disorders, bypassing damaged neuromuscular pathways. However, most existing BCIs are designed for specific applications. They capture neural activity from a specific brain region associated with a given task, extract key features from the recorded signals, and translate them into the user’s intent. The decoded information is transmitted as commands for discrete goal selection or continuous control of assistive devices. Even within the same application, BCI technologies vary significantly depending on the user’s communication and control capabilities. As a result, the current BCI ecosystem is fragmented and lacks flexibility, necessitating a more adaptable solution that supports multiple applications simultaneously. This research aims to address these limitations by introducing a ubiquitous BCI (uBCI) framework — a versatile system designed to enhance both cognitive and motor functions through advanced neural signal processing and an energy-efficient digital architecture. Unlike conventional BCIs, which rely on a single brain region and predefined features, the uBCI system analyzes signals from multiple brain regions, leveraging diverse neural features for real-time, accurate decoding of user intent. The uBCI framework enables both discrete goal selection and continuous control, ensuring long-term reliability in neural signal interpretation for practical applications. Beyond its scientific contributions, this research has the potential to transform the lives of millions of people affected by neurological disorders. Educational initiatives include the 'Brain Chips' outreach program, which engages high school students; the integration of research findings into the Computer Engineering curriculum at San Diego State University; open-access dissemination of research results; and community engagement through workshops at the Disability Center San Diego. The uBCI framework is built on a robust technological foundation that ensures adaptability, efficiency, and seamless integration across multiple applications. The system integrates a custom digital integrated circuit for in vivo processing with a programmable processor featuring a heterogeneous architecture for in silico processing. To evaluate efficiency, the system's components are implemented and tested on field-programmable gate arrays (FPGAs), communicating wirelessly via Bluetooth. The uBCI is validated and optimized using neural datasets from human and non-human primates, along with prosthetic testing on the humanoid Baxter Robot System. By optimizing neural signal processing algorithms and associated digital circuits across in vivo (biological) and in silico (computational) domains, this project advances fundamental BCI research while bridging the gap between laboratory prototypes and real-world deployment, paving the way for scalable and user-adaptive neural interfaces. 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 Awards · FY 2025 · 2025-06

Scientists assess potential earthquake hazards along known faults by studying past earthquakes that have occurred along those faults. Because earthquake recurrence intervals may be hundreds to thousands of years for any given fault, the number of earthquakes directly observed on that fault in recorded history is typically one or two at most. Paleoseismologists (geologists who study past earthquakes) can dig further back in time by looking at the geologic record of ancient earthquakes: evidence of past slip events that have disturbed rocks and sediment around the fault. This is an expensive and time-consuming process that involves excavating parts of a fault, mapping out geologic structures, and determining the absolute dates of materials disturbed by past ruptures. An alternative approach is to study indirect evidence of past earthquake ruptures preserved in the fractured and pulverized rocks surrounding faults in the so-called fault “damage zone”. A fundamental question in this research project is whether fault damage zones contain information about the maximum earthquake size a fault can host (Mmax) in terms of type, style, extent, width and degree of damage. This project will develop criteria to distinguish damage related to earthquake rupture from damage accrued over the longer-term growth of the fault, and to use these criteria to test the hypothesis that the style and intensity of damage on faults that experience earthquake magnitudes greater than ~Mw6.6 to 6.8 can be clearly distinguished from that of faults experiencing smaller magnitude events. This project will include a combination of field-based structural geology of crustal scale faults in southern California, cutting edge rock mechanics experiments, and theoretical rock and fracture mechanics to provide a roadmap for identifying uniquely seismic features preserved in damage zones, and to test the overarching hypothesis that the maximum earthquake size a fault can host can be estimated by examining the damage zone structure of active strike slip faults. Mmax is a critical component of probabilistic seismic hazard assessment (PSHA) as it limits the maximum size of earthquakes considered in a seismic hazard model. Slip rates of faults are the main drivers of hazard, but Mmax controls the upper end of moment release. If Mmax is large, then a significant proportion of the long-term seismic moment release is accommodated by rare large earthquakes. In PSHA, this decreases hazard, compared with moderate earthquakes, which also generate strong shaking but have higher recurrence rates. Hence, quantitative information on Mmax is a significant aspect of quantifying hazard to critical facilities. Current approaches for determining Mmax can be strengthened by developing independent criteria that allow for Mmax determination without knowing the full paleoseismic history. This research will lead to the development of criteria to distinguish damage related to earthquake rupture from quasi-static damage accrued over the longer-term fault evolution, and to use these criteria to test the hypothesis that the style and intensity of damage on faults that experience earthquake magnitudes greater than ~Mw6.6 to 6.8 can be clearly distinguished from that on faults experiencing smaller magnitude events. This will be accomplished by carefully documenting the shallow expression of damage zone structure of crustal scale faults in Southern California and examining the unique characteristics of brittle damage that varies as a function of known historical and paleoseismic earthquake magnitudes. The difference in damage state is likely to be explained by increased energy dissipation by off fault deformation above a critical moment magnitude threshold, and understanding this relation yields the potential for estimating Mmax for active faults with incomplete historical and paleoseismological records. This collaborative study will use field-based structural geology, cutting edge rock mechanics experiments, and theoretical rock and fracture mechanics to provide a roadmap for identifying uniquely seismic features preserved in damage zones, and to test the overarching hypothesis that Mmax can be estimated by examining the damage zone structure of active strike slip faults. This project has the potential for developing an independent, deterministic criterion for Mmax on individual active faults by examining the damage zone structure. 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 Awards · FY 2025 · 2025-06

This project aims to serve the national interest by mapping national patterns in the use of research-based instructional practices in post-secondary chemistry, mathematics, and physics courses five years after the disruptions due to the COVID-19 pandemic. In the Spring of 2019, a survey was sent to roughly 18,000 instructors of first-year mathematics, chemistry, and physics courses at nearly 1000 post-secondary institutions. That survey provided a comprehensive view of introductory science courses and instructors across the United States, with responses from nearly 4000 faculty from 660 U.S. colleges and universities. However, just one year later colleges and universities across the nation quickly shifted to online, emergency remote teaching in response to the COVID pandemic. The scale of instructional change during this time was both unprecedented and ubiquitous, with nearly every instructor teaching in the spring of 2020 required to try something new, and many needing to continue experimenting and revising their courses for the following semesters. This project will repeat the 2019 survey in order to characterize any lasting impact of the COVID pandemic on undergraduate science education and understand what this new instructional landscape may mean for change agents working to improve undergraduate science education through the uptake of research-based instructional practices. The goals of this project are to 1) understand the impact of the COVID pandemic on undergraduate science education as well as provide a current description of undergraduate science instruction, and 2) in consideration of any shifts following the COVID disruption to higher education, revise and update the research-based insights and recommendations for supporting and achieving instructional change in undergraduate STEM. To do so, the roughly 18,000 instructors will be re-surveyed. Some of the survey analyses will be conducted on the new responses alone, including multilevel modeling of the impact of malleable factors on instructors’ adoption of research-based instructional practices. Other analyses will incorporate the prior results for pre-post analysis to capture changes in the practices of both individuals and the disciplines in the aggregate. Where changes are observed, additional statistical tests and modeling will be used to identify the impact of emergency response teaching strategies on those shifts. These findings will be used by change agents (e.g., professional development organizations, instructional coaches) to better support undergraduate instructors in implementing research-based instructional strategies and by administrators (e.g., department chairs, course coordinators) in making resource allocations and policy decisions. These results will update the foundational knowledge base needed to support widespread pedagogical shifts toward the use of research-based instructional practices in post-secondary STEM education, impacting undergraduate students across the country. The NSF IUSE: EDU Program supports research and development projects to improve the effectiveness of STEM education for all students. Through its Institutional and Community Transformation track, the program supports efforts to transform and improve STEM education across institutions of higher education and disciplinary communities. 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 Awards · FY 2025 · 2025-04

This Faculty Early Career Development (CAREER) award supports research to investigate an innovative, cooling-assisted additive manufacturing (AM) approach for building bioinspired materials. By learning from nature's complex designs, bioinspired hybrid materials combine lightweight properties with exceptional strength, toughness, and impact resistance, making them ideal for applications in sports, biomedical engineering, and consumer electronics. Traditional bottom-up assembly and AM technologies face limitations in fabricating bioinspired materials with high ceramic loading, precise crystal size, and controlled distribution within polymer frameworks, which limit their functionality. This project addresses these challenges by developing a novel AM technique that integrates photo-induced polymerization and layered cooling crystallization to mimic biomineralization process. Success of this project will enable precise manufacturing of rigid materials in specific regions of polymer frame and enhance material functionality. Educational activities will engage students at all levels across the U.S. and prepare a highly skilled workforce. These efforts will strengthen the America’s talent pipeline, enhance public engagement, and equip future professionals to meet critical economic and societal challenges. This CAREER project aims to understand the processing mechanism of a new multi-material AM technique to produce dual-material properties using a single resin, eliminating post-processing while enhancing scalability and efficiency. The proposed research introduces a layered cooling crystallization AM approach to fabricate bioinspired hybrid materials by integrating programmable crystal growth within photopolymerized polymer frames. The project seeks to advance scientific comprehension of the intricate effect of cooling parameters (temperature, cooling time, and rate), resin chemistry (composition and supersaturation level), as well as bioinspired polymer patterns (structures, wettability, and roughness) on the nucleation, growth, crystallization dynamics and bonding of crystals during the layered cooling crystallization. Advanced simulations and experimental methods will elucidate the relationship between thermal fields, crystal growth kinetics, material properties, synergistic system integration, and resulting functionalities of the printed hybrid materials. By mimicking bio-growth processes through multi-material formation mechanisms, this approach will enhance design flexibility and lead to optimized manufacturing practices. The findings will provide a robust framework for designing multifunctional materials with applications in sports gear, smart protective devices, biomedical implants, and beyond, advancing both fundamental understanding and industrial capabilities in AM. 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 Awards · FY 2025 · 2025-04

This Faculty Early Career Development (CAREER) project supports the nation's research priority in climate adaptation through advancing scientific understanding of urban and coastal floodings and establishing a new generation of intelligent flood early warning systems and smart flood control infrastructure. The project will address gaps in flood data availability and in present-day flood modeling by harnessing heterogeneous data sources, such as ground-based images and videos taken by traffic cameras, smartphones and drones, to provide faster and more distributed flood timeseries data and visual information. The use of multi-source, heterogeneous data, along with accelerated flood modeling and data analytics, will support the transformation of existing flood infrastructure into Active Flood Control Infrastructure through real-time model updating, active measurements, and active flood management. The project integrates research activities with educational and outreach plans to (i) train the next generation of AI-enabled engineers and scientists, (ii) foster flood-aware communities, and (iii) inform decision-makers by developing an integrated looped learning framework consisting of four phases of flood modeling, planning and response, behavioral analysis, and virtual reality gameplay and outreach. The project will study two coastal watersheds in South Carolina, one dominantly urbanized and the other natural, with potential transferability to other urban and coastal systems. Novel Artificial Intelligence and image processing tools will be deployed to process different types of inputs at different stages of flood management: data acquisition, flood detection, monitoring, simulation, and forecasting. The research core of this project revolves around the detailed design and development of an integrated modeling platform consisting of Multi-Deep Learning Models (MDLM) for flood data analysis, modeling, and management. In the data analysis phase, deep learning models, alone or in combination with the reconstruction of a 3-Dimensional of the study area, will be used to provide numerical data, such as water levels, and inundation area, from ground-based images, and videos. Moreover, a multi-source data fusion module will be developed to feed data from different satellites and bands to a multi-branch deep learning network for flood detection and feature extraction. In the modeling phase, this CAREER project will enhance the understanding of compound flooding by developing a fully coupled model for simulating coastal compound floods through the integration of distributed hydrologic and surface hydraulics mathematical models into a single modeling framework. Then, a set of machine learning-based surrogate models will be developed to mimic the knowledge of fine-scale physics-based flood models and provide timely flood predictions. Finally, this project will provide adaptive design guidelines to turn existing infrastructure into active flood control infrastructure through real-time model updating, active measurements, and active flood management. New technologies and tools created in this project will allow stakeholders, decision-makers, and the public to make choices regarding their direct and indirect involvement pre-, peri-, and post-flooding and evaluate their impacts using integrated numerical simulations and virtual reality gameplay. This CAREER project is jointly funded by the Civil Infrastructure Systems (CIS) and the Established Program to Stimulate Competitive Research (EPSCoR) 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 Awards · FY 2025 · 2025-03

Bacteriophages are viruses that infect bacteria. They are the most numerous life-forms on the planet. In general, bacteriophages consist of a genome made either of DNA or RNA that is packaged into a protective protein shell called a capsid. While much is known about bacteriophages with DNA genomes, far less is understood about those with RNA genomes. This project will investigate how these RNA genomes are packaged into their protein capsids. In particular, the project will describe the shapes into which the RNA genomes are folded, and how these shapes are recognized by the proteins and selected for packaging. By identifying these structures, the project will inform the design of synthetic capsids capable of packaging synthetic RNA molecules for use in RNA-based technologies. Furthermore, these RNA structures will facilitate the discovery of new RNA viruses by learning how to recognize viral RNA in the sequence libraries obtained directly from the environment. These discoveries will advance the understanding of RNA bacteriophage diversity and enable the production of novel nanoscale materials derived from their coat proteins. Throughout, the project will train graduate and undergraduate students in challenging fields of physical chemistry, biochemistry, and computational biology. Bacteriophage MS2 is the canonical model system for studying viral RNA packaging, yet the mechanisms by which MS2 coat proteins selectively package their RNA genomes while excluding abundant cellular RNA remain unclear. While RNA of a model phage MS2 is believed to contain structural signals that are recognized by its coat proteins during packaging, the precise nature of these signals and their role in selective packaging is not fully understood. This project will use quantitative packaging experiments inside the cells to identify key RNA structures involved in selectivity. Plasmid-encoded RNA molecules with defined structures will compete with the cellular transcriptome for packaging by MS2 coat proteins. Single-particle imaging and high-throughput sequencing will quantify the packaged RNA composition, providing a direct measure of selectivity. These measurements will be used to critically evaluate current hypotheses of selective packaging and identify the key RNA structures involved. The project will then use these structures to screen metagenomic datasets for previously undescribed RNA bacteriophage genomes. Once discovered, the coat proteins from newly identified bacteriophages will be expressed in cells to produce protein capsids. In addition to verifying the discovered bacteriophages, these capsids will provide the research community with new protein nanoparticles for applications in biotechnology and nanoscience. 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 Awards · FY 2025 · 2025-03

The project aims to further the understanding of collapse-type phenomena and rogue waves in systems that are modeled by nonlinear ordinary, partial, and lattice differential equations. Collapse-type phenomena are mathematically described by solutions that remain self-similar as some of their attributes become unbounded in finite time. Self-similarity refers to preservation of shape when an appropriate scaling of space and time and solution amplitude is employed. Collapse phenomena are relevant to the focusing of light beams in optics and to atomic matter waves. Rogue waves have a characteristic length or time scale and extreme amplitudes; they are important in subjects such as hydrodynamics, nonlinear optics, and atomic and plasma physics. Using dynamical systems and computational techniques, this project aims to reformulate the underlying models and provide a unified approach to studying both collapse phenomena and rogue waves by treating the relevant patterns as self-similar solutions. The project is expected to provide insights on the mechanisms and reduced mathematical descriptions of collapse phenomena in some of the prototypical mathematical models that feature these potentially catastrophic focusing events, as well as on the formation, prediction, and analysis of extreme waves in both continuum and spatially discrete systems. The project will offer research training opportunities for students. The project will explore a recently derived normal form for the study of self-focusing waves of the central dispersive wave model of the nonlinear Schrödinger equation and will seek generalizations for related models (such as the Korteweg-de Vries equation). Stability analysis of such collapsing waves is expected to shed light on the spectral properties and potential instabilities of such systems, their connection to symmetries, their implications for the dynamics in different settings (supercritical, critical, and subcritical), and their reinterpretation in the original "non-exploding'' frame. A second focus of the project will be the study of rogue waves from a dynamical systems viewpoint, including characterization of stability via limits of time-periodic solutions and rogue waves in higher-order dispersion settings. The theoretical analysis will be corroborated by numerical simulations involving deflation-based fixed-point techniques and pseudo-arclength continuation, as well as state-of-the-art contour integral-based eigenvalue solvers. Collaboration with experimental groups performing laboratory experiments will also be sought. 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 Awards · FY 2024 · 2024-11

This project will contribute to the national need for well-educated scientists, mathematicians, engineers, and technicians by supporting the retention and graduation of high-achieving, low-income students with demonstrated financial need at San Diego State University (SDSU), a Hispanic Serving Institution, partnering with Imperial Valley College (IVC), also a Hispanic Serving Institution. Over its one-year duration, investigators at these two institutions will develop plans to fund students who are pursuing associate’s, bachelor’s, and graduate degrees in electrical, computer, and cyber-physical systems engineering disciplines. The eventual goal is to provide 200 scholarships to eligible students. The project will build on prior work implementing effective support for underrepresented students in STEM, providing additional insights within the context of a preliminary needs assessment, and evaluation of the pilot activities as a whole. Based on this collaborative planning an anticipated Track 3 project will propose to study the impact of new engineering degrees on local low-income students. The project would focus on low-income students and their intersections with other historically underrepresented populations in STEM fields, particularly Hispanic and first-generation students. Collaboration among SDSU and IVC faculty and program staff will provide the additional supports that these students need to persist in STEM, including cohort building, faculty mentorship, near-peer role models, and networking opportunities. The project will lead to increased opportunities for students with the greatest need, ultimately leading to increased diversity within the STEM workforce. The overall goal of this project is to increase STEM degree completion of low-income, high-achieving undergraduates with demonstrated financial need. The vision for this project is to create a cohesive plan towards a Track 3 proposal that will support eligible students at the partner institutions. To achieve this vision, the project will have three core objectives: (a) identifying the senior personnel and offices to contribute to the Track 3 proposal; (b) building effective collaborations to complete the needs assessment, articulation agreements, and scholarship targets; and (c) developing a future Track 3 proposal. The project will use pilot activities including cohort building, transfer and graduate school preparation, and participation in discipline-specific conferences to gather data and evidence for developing interventions to support scholars. This project is funded by NSF’s Scholarships in Science, Technology, Engineering, and Mathematics program, which seeks to increase the number of academically talented, low-income students with demonstrated financial need who earn degrees in STEM fields. It also aims to improve the education of future STEM workers, and to generate knowledge about academic success, retention, transfer, graduation, and academic/career pathways of low-income students. 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 Awards · FY 2024 · 2024-10

This project aims to serve the national interest by developing an empirically validated tool to assess a student’s knowledge of computing concepts. Such well-designed knowledge assessments, vetted by rigorous design and psychometric processes, are still relatively uncommon in the computer science education community. Moreover, students learn computing concepts in a variety of increasingly different educational contexts: at home, on the job, in a course offered online, or in a traditional classroom. Thus, there is a pressing need for an empirically validated assessment of computing knowledge that works in a variety of educational contexts and with different programming languages. The goal of this project is to design and pilot such a multi-contextual and multi-language assessment by enhancing the existing Second Computer Science 1 (SCS1) course assessment. The new assessment, referred to as SCS1++, will be designed by addressing known issues with the existing SCS1 assessment while also increasing the number of questions on the assessment. Specifically, the project will (1) construct a coherent argument for validity claims of SCS1++ as a whole; and (2) create subscales aligned with the concepts on the assessment to improve the assessment’s formative values. The project will directly inform the research on equitable CS assessment by updating the current SCS1 with improved questions and rigorous validation using advanced psychometric tools. A design-based research approach will be used, based on theory, and the system will be evaluated in real educational settings. The NSF IUSE: EDU Program supports research and development projects to improve the effectiveness of STEM education for all students. Through its Engaged Student Learning track, the program supports the creation, exploration, and implementation of promising practices and tools. 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 Awards · FY 2024 · 2024-10

Deep neural networks have led to significant advances in science and engineering and play an important role in the success of modern machine learning in various real-world applications including vision, speech, pattern recognition, and biology to name a few. When developing deep-learning solutions, accuracy or performance metrics are often a key point of emphasis. While performance is critical, the computational load of the training process and security of the final solution play an equally important role in a real-world setting. Recent advances in adversarial learning models hold significant promise in improving various learning methods and defending against threats, but the fundamental aspects of these models are still poorly understood, which limits their performance guarantees for efficient and robust decisions. With this in mind, this project investigates simultaneously tackling three desirable properties when developing deep networks: 1) performance, 2) efficiency, and 3) robustness. This project also includes a comprehensive plan to integrate the research results into cross-disciplinary educational multilevel programs by funding graduate research assistants, summer research fellowship for high-school students and teachers, and organizing a hybrid (online and in-person) deep-learning boot camp. The overall goal of this research program is to develop a comprehensive and fundamental understanding of the robustness and computational aspects of deep networks by leveraging tools and concepts from probability, information theory, and statistics. This project aims to make critical advances in 1) proper formulations of subnetwork adversarial robustness, 2) characterizing transferability via curriculum learning, and 3) developing efficient approaches for reducing computational complexity involved in training, among others. The theoretical and methodological outcomes of this cross-disciplinary project will broaden the prior knowledge of deep learning and will improve prediction, exploration, and detection applications of machine-learning models. 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.