Georgia Tech Research Corporation

NSF Awards · FY 2025 · 2025-07

A great number of problems in physics, mathematics and other scientific disciplines are variational, that is, their solutions are obtained by optimizing cost or energy functions subject to certain constraints. A classical question, for instance, is concerned with the most efficient way to move resources from one place to another. The mathematical theory addressing this is the theory of optimal transport. It has deep connections to analysis, geometry, and probability, and plays a central role in a wide range of applications, including economics, fluid dynamics, and machine learning. While the existence of optimal transport plans can be established in great generality, the study of their structure and fine properties is more delicate. This is the aim of regularity theory, which analyzes whether solutions are well-behaved or develop singularities. These insights are important for theoretical developments and practical applications. The investigator studies the regularity of optimal transport plans in situations relevant to quantum chemistry and data science. There are also connections to materials science: regularity theory turns out to be related to the mathematical understanding of microstructure in materials such as thin-film ferromagnets and superconductors. Understanding the emergence of complex patterns has been a long-lasting effort by scientists. From the mathematical point of view, proving their existence and formation is an extremely challenging task. This project has a broad educational impact by integrating cutting-edge mathematical ideas into both graduate and undergraduate education. It provides valuable training opportunities for graduate students through direct involvement in research-level mathematics. The focus of the project is the regularity of optimal transport. This concerns classical as well as multi-marginal optimal transport, which addresses the question of correlating more than two measures in the most efficient way. Until recently, the regularity theory of optimal transport has mostly been based on partial differential equations techniques. However, one can also obtain regularity results for optimal transport plans based on energy estimates, that is, within a fully variational framework; this approach is orthogonal to the classical one based on Monge-Ampère equations and is similar to De Giorgi’s approach to the regularity of minimal surfaces. The investigator develops this variational approach further, in particular in degenerate and singular situations such as optimal transport with Coulomb cost, which is of relevance in electronic density functional theory – a central tool in the investigation of quantum systems and the development of new quantum technologies. The investigator also explores the regularity of multi-marginal optimal transport and Wasserstein barycenters. A different but closely related research direction concerns the development of a robust mathematical framework to understand and analyze branching phenomena, focusing on the description of branched microstructures in a statistical sense. On the mathematical level, this can be seen as regularity theory in action. 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

Cyber-physical systems (CPS) consist of computational nodes tightly integrated with their physical environments. Such systems often exhibit repetitive or periodic behaviors, for example, orchestrated leg movements in a robotic dog. Other examples include multi-finger manipulators, traffic flow control, and chemical plants. Synthesis of such periodic behaviors is a challenging problem since computation quickly becomes intractable as system complexity increases. This project will alleviate such computational challenges and ensure that synthesized motions are robust against uncertainties in the system model and in its environment, as well as external disturbances. The technical approach builds upon recent advancements in contraction theory and will extend this theory to cover hybrid behaviors (i.e., behaviors that have both a continuous-time evolutions and instantaneous discrete jumps). The work will focus on providing interpretable and scalable guarantees of contraction, which in turn will ensure robustness of behaviors. This methodology will leverage the natural dynamics of the system and decrease the required energy consumption for stabilization. The project is decomposed into three separate contributions: (1) the development of a theoretical foundation of contraction in hybrid systems to approximate regions of attraction for contact rich CPS; (2) the design of computation-aware algorithms for efficient computation of the regions of attraction for practical motion synthesis; and (3) the experimental validation of the entire framework on various contact-rich mechanical CPS, including bipedal and multi-fingered robots. 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

This project aims to serve the national interest by establishing practices to improve mentorship, learning, and engagement in undergraduate research experiences (UREs) in engineering. UREs play a crucial role in enhancing engineering students' academic experiences by providing hands-on, authentic experiential learning opportunities. This level 2 project in the Engaged Student Learning track of the IUSE program will build on an existing toolkit of interventions and best practices that create structured reflection and mentoring activities within engineering UREs. The implementation, adaptation, and outcomes of this toolkit in UREs will be studied across five institutions. This work has the potential to expand access and improve the quality of mentorship in engineering UREs. It will also strengthen the existing partnership between a diverse group of participating institutions, and the knowledge gained will result in the expansion of high-quality, freely available resources for engineering UREs. During the project, the investigators will form an Undergraduate Research Excellence Network (UREN) with URE mentors, including faculty, postdoctoral researchers, and graduate students, to implement the existing toolkit. The toolkit includes student and mentor training videos, activities, instructor guides, and workshops that make the learning and benefits of UREs more visible to and accessible for students and faculty. The UREN will provide structured training and coaching to help mentors form an action plan for adopting these tools in their research mentorship. The implementations will be evaluated through an emergent design studies approach, focusing on mediating processes involving student self-assessment, reflective thinking, and mindset. These findings will be linked with evaluated outcomes related to students' autonomy and self-confidence in research and their ability to connect their research experience with their learning through coursework. This project will increase knowledge about improving learning and engagement in engineering UREs. Evaluating whether the learning gains observed in previous work can be generalized across five institutions will establish this toolkit as an important contribution to enhancing engineering students' learning. Furthermore, it will provide critical insight to inform the development of new pedagogical approaches that leverage student mindset to support experiential learning in STEM fields. The NSF IUSE: EDU Program supports research and development projects to improve the effectiveness of STEM education for all students. Through the 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 2025 · 2025-07

Security of most practical cryptographic protocols relies on secret keys and hence is subject to brute-force (exhaustive search) attacks. This means that an attacker can attempt to test all possible keys in order to find the right one. This is usually not a big concern, because the exhaustive search of strong cryptographic secret keys is infeasible (i.e., it takes millions of years). However, in practice, keys are often generated from users' passwords or biometric data (such as fingerprints or facial features) for convenience of key management. In this scenario, brute-force attacks become a big concern because testing all passwords and biometrics is often feasible; for instance, it was recently shown that fingerprints used to unlock phones could be brute-forced in minutes. While there are some solutions for protecting against brute-force attacks of passwords, it is not straightforward how to adapt such solutions to biometrics. This is an important problem because biometrics are increasingly used in real systems and, unlike passwords, biometrics cannot be easily changed. This project's goal is to develop new techniques and protocols to protect sensitive biometric data against brute-force attacks. The focus of this project is solving the most pressing open problem in the security of highly sensitive biometric data. This project will develop a novel protocol for biometric-based authentication and key reconstruction that is highly resistant to brute-force attacks. In order to provide the security guarantees, the project will specify a security model capturing very strong adversarial capabilities and prove security of the protocol according to the security model. An important part of the project is to implement the protocol for concrete biometric datasets and tune its parameters for acceptable precision and security levels. Finally, the project will extend the solution to a brute-force-resistant, biometric-based authenticated key exchange protocol. 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

Online authentication is crucial for protecting users' accounts and data online. For decades, online authentication has relied upon passwords. However, passwords have widely-recognized security and usability issues. These concerns have driven the creation of passwordless authentication approaches, including promising methods like Fast IDentity Online 2 (FIDO2) that have already seen notable initial deployment by popular browsers, devices, and online services. FIDO2 uses cryptography instead of passwords to support authentication, and offers desirable security and usability properties. Given that these technologies are still relatively new and quite different from password-based authentication, it is unclear how both users and online services will act as they move toward using passwordless authentication, and what unexpected security and usability issues may arise. This project seeks to understand how passwordless authentication will manifest in practice, as well as the emergent security or usability problems with it. The project will also produce solutions for addressing these concerns, which are particularly important to develop before passwordless authentication is more broadly deployed. The project also includes a substantial educational effort to introduce authentication-related computer security topics to high school students. Ultimately, this project will expand the knowledge of online authentication and security, and support the transition to a secure and user-friendly passwordless future. As authentication is fundamentally a human-driven task, this project will apply a hybrid approach combining user-driven studies, web measurements, and system design. The technical aims of this project are divided across three integrated tasks that consider both end users and online services. The first task will systematically investigate passwordless adoption dynamics, monitoring real-world adoption behavior and illuminating socio-technical barriers to adoption. The second task will evaluate the security and usability implications of real-world passwordless behavior once adopted, particularly at scale (considering passwordless use across multiple devices and online services, as well as by all user populations). The third task will systematize the insights gained from the earlier tasks, producing system and design principles for passwordless authentication. It will then apply these principles to develop socio-technical solutions to address emergent adoption, security, and usability issues with passwordless authentication, and improve its use in practice. 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

Data visualization is a powerful tool commonly used by practitioners to understand, analyze, and identify interesting patterns in their data. However, due to limitations in human perception and issues of efficiency in data processing and rendering, it has become standard practice for data analysts to work with only a sampled subset of their data. However, deriving samples that preserve key aspects of the underlying data is not straightforward. Analysts looking at a sample should be able to draw the same conclusions as they would from the complete original data. Unfortunately, this is often not the case. For example, while uniform random sampling is the most accessible method to data analysts in practice, it frequently fails to represent outliers and patterns, leading to missed critical insights and distorted perceptions of the underlying data trends. This is not a rare occurrence, as interesting insights often lie within outlier behaviors, edge cases, or uncommon aspects of the data. This is detrimental to visual data analysis: sampling choices can critically impact the accuracy and efficiency of common visual analytics tasks, and, when chosen poorly, can lead analysts to incorrect conclusions. This project addresses the gap at the core of this issue: data systems do not model human perception, and sampling algorithms do not optimize for it. Explicitly accounting for human perception in samples can create more effective visualizations and reduce the risk of distorting patterns and trends in the data. Considering that the data analytics market generated revenue of almost US$23 Billion in 2019 and is projected to reach US$132 Billion by 2026, improvements in the visual analytics domain are positioned to have a tremendous impact on the economy. As decisions often rely on visual analysis, improved accuracy in inferences will lead to better decision-making and data communication, which can also support data-driven decision-making by policy-makers and improve the interpretability and credibility of data analyses for the general public. This project will augment data management systems with explicit models of perceptual features of the data, and will contribute algorithms that use perceptual models in data selection for visualization tasks. The investigators will explore the suitability of established saliency models, commonly used in computer vision, that predict areas of a visualization that attract viewers' attention as a proxy of perception. The project will make the following intellectual contributions: (1) A prototype perception-augmented database, with novel representations for perceptual saliency data and optimized storage strategies to minimize overhead. (2) Novel sampling algorithms that use perceptual models toward their data selection objectives, including optimizations for on-the-fly sample augmentation and robustness across a range of visualization tasks. (3) Perception-aware compressed representations of data that can be used towards approximate visualizations for increased efficiency and real-time performance. (4) Novel measures for perceptual quality in samples. These advances will enable data scientists to draw accurate insights from visual analysis, within an efficient and robust analytics pipeline. 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

Science and engineering research has evolved to increasingly rely on data-intensive and artificial intelligence (AI) driven methodologies. This, in turn, has led to more complex workflows that involve hardware and software resources at different scales and operating in different environments. These growing complexities necessitate a flexible and heterogeneous hardware and software infrastructure that transforms existing, disjointed workflows via a seamless and more productive computational environment that is easier and more intuitive to use. Nexus provides such a critical national resource to the science and engineering research community; it serves both as a standalone platform and as a gateway to effectively utilizing other national resources, significantly accelerating AI-driven scientific discovery. Nexus will advance American leadership in artificial intelligence and advantage U.S.-based research and industry enterprises in diverse areas of science and engineering, enabling breakthrough discoveries, increasing economic competitiveness, and advancing human health. Nexus is a next-generation, national-scale computational resource integrating cutting-edge heterogeneous hardware, AI-accelerated computing, and advanced software services to unify scientific and engineering research workflows. Led by the Georgia Institute of Technology in collaboration with the National Center for Supercomputing Applications (NCSA) at the University of Illinois Urbana-Champaign, Nexus provides a seamless and scalable computational environment that bridges local and national resources. The Nexus computational infrastructure incorporates substantial CPU and GPU resources that deliver 14.5 petaflops of peak double-precision calculations and over 400 petaflops of peak AI-focused, reduced-precision calculations. This computing capability is augmented by high-speed memory, a 10-petabyte all-flash file system, and a next-generation high-speed network based on InfiniBand and Socket Direct technologies. This infrastructure is integrated with a novel software platform incorporating the Nexus computing infrastructure within researchers’ local environments, allowing them to harness powerful computational capabilities without disrupting their workflows. The Nexus platform will also facilitate federated resource sharing, enabling interoperability with national cyberinfrastructure assets, including the NSF-funded Delta and DeltaAI systems at NCSA. Finally, it supports multi-scale, heterogeneous computing, giving researchers access to diverse computational architectures through a unified interface, an essential capability for AI-driven research and computational sciences that require tailored, scalable computing power. Regional engagements, training workshops, student and faculty development programs, and collaborative research partnerships will be developed to leverage innovative features of the Nexus resource for supporting and expanding the nation’s globally competitive STEM workforce equipped with essential AI and HPC skills. 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

Service robots can be of great help in hospitals, caring for older adults, and for everyday home tasks. However, today's robots need a large amount of retraining whenever they face new situations. This is like having to go back to school for every new job. This Faculty Early Career Development (CAREER) project research will help robots learn like humans do - learning on the job. The new system will enable home robots to learn new skills from human teachers quickly and on their own. It will also allow them to adapt fast to new tasks through smart planning. To get young students excited about robotics and artificial intelligence (AI), the team will develop teaching programs for K-12 students from different backgrounds. The project will also create new college classes that combine AI and robotics to prepare students for future jobs. Robot learning on its own is challenging in real world scenarios - robots must learn individual skills as parts of longer, more complex tasks. For example, learning to cut vegetables requires first learning to grasp knives well. This research will develop a new robot system where intelligent planning helps guide the robot to learn. The system helps the robot find missing skills and design situations that help the robot to practice these skills. The system also uses these learned skills to help the robot plan for more complex tasks. The system has three main components: (1) learning new skills with a symbolic planner as guidance; (2) efficient learning from occasional human guidance when needed; and (3) fast adaptation to new tasks through intelligently combining learned skills. The team will test the system in simulated environments and on real robots used in home care research centers. The focus will be on daily household tasks that need long-term adaptation. This research will speed up the adoption of capable robots in healthcare, manufacturing, and home assistance areas. 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

Developing neighborhood connectivity, trust, and belonging is essential for the mental health of young people. Social connections help mitigate the stressors young people face. Technologies that build community engagement and include information of public and local interest, that is, civic technologies, have shown promise for improving social connections. Yet existing tools often fail to sustain public interest or help the public intuitively navigate the complex data displayed, and also often lack transparency. Little is known about how civic technologies can be designed to engage young people, whose ideas about what community means and how it is best formed, often differ from the adult-centered views driving existing tools. Addressing these research gaps, the project team is creating and evaluating a toolkit that will include a novel suite of data and technology to support youth-led civic engagement, build a more socially cohesive community, and in turn support youth mental health and the wellbeing of people in the local community. Through collaborations with community and municipal organizations, this work is improving civic data access for youth and their communities, enabling community-grounded data interactions that advance social cohesion and wellbeing, and helping build data literacy, an essential 21st century skill. This research is engaging youth in the co-design of a toolkit and activities for building data literacy, software tools for community data collection, visualization, and storytelling, and templates to represent data in physical forms that enable learning, communication, problem solving, and decision making. Mixed-method studies are evaluating how using these community tools can support key dimensions of data work (e.g., collecting, analyzing, and representing neighborhood wellbeing data), youth dialogue around data insights with their communities, data transparency, civic engagement that catalyzes neighborhood cohesion and, ultimately, young people's resilience to stressors. While smart city innovation has accelerated, it rarely centers on the creative vision and agency of youth. Addressing that gap, this work is creating and testing a novel community data platform that helps youth achieve their aspirations for improved wellbeing in their neighborhoods. This empirical research is yielding previously unavailable evidence regarding patterns of adoption and engagement with civic technologies amongst youth, and how such tools can be designed to support youth-led, community-grounded, and data-driven interactions that strengthen neighborhoods. 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

The ability to accurately predict multiscale fluid dynamics phenomena is essential for accurate weather forecasting, climate modeling, design of fusion energy devices and numerous other practical applications. However, the predictive modeling of multiscale phenomena is a challenging problem. Computer simulations solve the relevant equations at specific points that form a grid. Simulations on coarse grids can be made today, but they require accurate models of physics below the scale of the computational grid, i.e., between points on the grid. No general and systematic approach to construction of such sub-grid-scale models currently exists to capture all of the flow physics. The objective of this project is therefore to develop a systematic approach to modeling multiscale phenomena at a desired level of resolution. The fundamental advances resulting from the project will also impact a range of other applications in science (e.g., dynamics of accretion disks, forest/brush fires, pollution transport) and engineering (e.g., aircraft design, combustion and hypersonic vehicles). The project will also educate and train several graduate and undergraduate students in state-of-the-art numerical modeling techniques supplemented by novel machine-learning architectures, providing them with skills directly transferable to jobs in research and development, applied engineering applications, and national security applications. Traditional, analytic coarse-graining approaches for modeling multiscale problems often lead to infinite hierarchies of equations, with numerous ad hoc assumptions employed to close the truncated system of equations. The resulting models lack generalizability and are fundamentally incapable of correctly describing fluxes of physical quantities from small scales to large scales. Models constructed using deep learning, on the other hand, lack the interpretability needed to understand the mechanisms responsible for generating structure at multiple scales. This challenge will be addressed by developing a systematic machine learning framework combining relevant, limited domain knowledge and plentiful, representative, and well-resolved data generated using experiments and/or numerical simulations. This new framework will allow construction of explicit mathematical models of multiscale phenomena without relying on empirical assumptions that may or may not be justified in particular geometries or under particular conditions. The explicit analytical form of an inferred model will allow straightforward interpretation of various terms and enable identification of the dominant mechanisms responsible for transport of physical quantities of interest between scales. 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

Neutrinos are unique messengers, carrying information about the universe's most energetic astrophysical phenomena. Over the past decade, the IceCube Neutrino Observatory at the South Pole has made key discoveries by detecting high-energy neutrinos and identifying two active galaxies as neutrino sources. However, sub-TeV neutrinos (10–1000 GeV) remain a largely unexplored frontier with the potential to significantly expand our observation of the universe. This project leverages advanced artificial intelligence (AI) techniques to overcome computational challenges and improve the reconstruction of sub-TeV neutrinos using IceCube’s DeepCore subdetector. These advancements will enable detailed studies of astrophysical sources such as NGC 1068 at lower energy scales and pave the way for real-time public alerts of sub-TeV neutrinos, fostering coordinated follow-up observations across the electromagnetic spectrum. In addition to advancing neutrino astrophysics, the AI methodologies developed here will benefit a wide range of fields with similar challenges, including weather forecasting, neuroscience, smart cities, and precision farming, by enhancing the analysis of distributed sensor data. By integrating education initiatives, outreach programs, and undergraduate participation, this project promotes access to advanced research, supports STEM, and inspires the next generation of scientists and engineers. This project addresses the computational barriers to sub-TeV neutrino reconstruction through four interconnected research tasks. First, it designs a novel AI architecture capable of managing spatial and temporal irregularities in IceCube’s sensor data while embedding physics invariance for improved accuracy. Second, the project enhances computational efficiency by leveraging physics-informed inductive biases, enabling real-time processing of millions of neutrino events with low latency. Third, robust training methodologies will be implemented to address systematic uncertainties. Finally, the project investigates approximate symmetry modeling to allow AI models to adapt to practical deviations from the ideal physical model without compromising performance. These innovations will significantly improve the angular resolution and sensitivity of DeepCore to sub-TeV neutrinos, paving the way for transformative discoveries about astrophysical sources such as Seyfert galaxies as well as objects in the Milky Way. The tools and methodologies developed are designed to be broadly applicable, enabling breakthroughs in other scientific domains that rely on the analysis of complex spatiotemporal data. 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

Meta-PURE: End-Use-Driven Cell-Free Modules Cell-free systems reduce biology to its most basic parts, simplifying the complexity of traditional biomanufacturing. However, scientists still do not have an efficient way to “plug-and-play” these individual cell-free pieces. This project addresses this gap by developing a modular, standardized platform for cell-free bioprocessing. Instead of custom, one-off designs, the project creates eight reusable modules covering essential functions like energy generation and protein expression. These “ready-to-use” modules make it easier to mix and match capabilities, allowing for faster, more efficient product development. The team will demonstrate how modular design and standardized kits can reduce costs, improve accessibility, and boost productivity across three market-relevant targets. In doing so, this project aims to make cell-free technologies broadly available and expand the use of cell-free technologies across the U.S. bioeconomy. This project will also cultivate the next generation of biotechnology talent through a dynamic, cross-sector team of postdocs and graduate students—uniting expertise from industry, academia, and government—to pioneer and scale the future of cell-free biomanufacturing. The project pioneers a modular, standardized framework to address bottlenecks for scaling cell-free bioprocessing. The project will develop, characterize, and integrate eight distinct functional modules common across cell-free systems: three for energy generation, three for the synthesis of valuable products, and two for transcription–translation. These modules span varying levels of complexity—from individual purified enzymes to enzyme cascades, PUrified Recombinant Elements (PURE), and lysate-based systems—encompassing the full spectrum of cell-free biotechnologies. This modularity allows innovation beyond standalone PURE technologies; it comprises modules with defined capabilities, each built and optimized independently, then integrated to achieve more complex and ambitious synthesis goals. The project also incorporates advanced analytical tools and establishes new standards and metrics to rigorously evaluate module performance and compatibility, which will result in a system that provides reliable synthesis. Further, the system will be validated through the synthesis of three industrially relevant targets: the valuable small molecule santalene; the GamS protein, which enhances cell-free protein expression; and production of a bacteriophage. These demonstrations will show how the system can be used for a varied of use cases. This project will accelerate innovation, enhance reproducibility, and demonstrate a replicable framework that enables flexible, cost-effective, and high-performance bio-based production platforms across diverse applications to support scalable biomanufacturing and strengthen the infrastructure of the U.S. bioeconomy. 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

The annual Symposium on Parallelism in Algorithms and Architectures (SPAA) is a leading computing forum for research on parallel and distributed algorithms, data structures, and models. Participation in SPAA is a valuable aspect of the graduate school experience that furthers student careers and outcomes. This project will enable United States-based student participation, helping them forge new connections and collaborations and thereby advancing computing research in societally beneficial ways, contributing to the development of a more competitive workforce. The scope of SPAA spans both theoretical and systems research in parallel and memory-efficient algorithms. The goal of the project is to support undergraduate and graduate students who currently study in US institutes to attend the SPAA conference to present their research and network with leaders in the community. The funds will be used towards travel and conference registration expenses to reduce the cost of attendance for students and help increase participation. Participation in SPAA will enable student development by (a) enabling students to work on their presentation skills and introduce themselves to the community, (b) providing students an opportunity to learn about the state of the art research on parallel computing in theory, and (c) providing students opportunities for meeting and getting advice from leading researchers across academia and industry. 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

Proteins play a central role in many processes essential to life and have wide-ranging applications in medicine, energy, agriculture, and biotechnology. However, natural proteins are often not ideal for these practical uses. Protein engineering, a field that aims to design proteins with improved or novel functions, has transformed industries by creating tailored proteins. While traditional approaches, such as the Nobel Prize-recognized directed evolution method, have been remarkably successful in numerous protein engineering applications, they are typically slow, costly, and resource-intensive. This project seeks to advance protein engineering by combining cutting-edge artificial intelligence (AI) methods with advanced laboratory automation. By harnessing the power of AI to predict and design protein sequences and integrating it with an automated experimental platform, this research aims to greatly accelerate the discovery of new proteins, offering immense potential across multiple scientific domains with significant commercial and societal impact on medicine, biotechnology, energy, agriculture, chemical manufacturing, consumer products, and more. This project introduces a novel interdisciplinary approach leveraging recent AI breakthroughs in large language models and generative models, to guide protein function analysis and protein engineering, unlocking an unparalleled efficiency for functional protein discovery. The research focuses on developing new AI techniques tailored to the unique challenges of protein engineering, such as sparse data and the need to balance multiple complex protein properties. By leveraging protein evolution insights and generative modeling, the AI system will guide the design of functional proteins with enhanced properties. An integrated automated Biofoundry will design, create and test AI-designed proteins, validating and then refining the design as needed, enabling a high-throughput, closed-loop discovery process. Beyond advancing the field of protein engineering, the project's algorithmic innovations will contribute to foundational research in AI and computing, with the potential for broad applications in other scientific and technological domains. This award is co-funded by the Directorate for Computer and Information Science and Engineering and by the Directorate for Biological 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 Awards · FY 2025 · 2025-07

This I-Corps project focuses on the development of a fast and accurate simulation and design space exploration platform for chip design and hardware acceleration. Modern chip design workflows rely heavily on simulation tools that are either fast but inaccurate, or accurate but extremely slow, making it difficult for engineers to evaluate and optimize complex designs. These difficulties create a critical bottleneck in the development of high-performance computing systems, leading to increased costs, longer time-to-market, and underutilized hardware. This solution addresses these challenges by enabling engineers to simulate and evaluate chip designs significantly faster while maintaining high accuracy. This advance has the potential to improve productivity for thousands of engineers, shorten development cycles for new technologies, and reduce infrastructure needs. In doing so, the technology supports national efforts to advance semiconductor innovation, enhance energy-efficient computing, and strengthen domestic design capabilities in a rapidly evolving technology landscape. This I-Corps project utilizes experiential learning coupled with a first-hand investigation of the industry ecosystem to assess the translation potential of the technology. This solution is based on the development of an intermediate representation–level simulation framework that provides cycle-level performance estimates without the need for slow and resource-intensive register-transfer level simulation. The approach decouples functionality from performance modeling, achieving over 99.9% accuracy with speed improvements of up to 200 times. The technology also supports rapid incremental exploration of hardware design parameters, such as memory partitioning, dataflow configuration, and communication buffering, enabling efficient architectural tuning. These capabilities are integrated into a modular and tool-compatible flow that benefits users by significantly accelerating feedback cycles and reducing dependence on manual edits or full synthesis. As a result, this solution empowers more scalable and informed decision-making in system design, helping researchers and engineers innovate faster and with greater efficiency. 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

The origin of multicellular life stands among the most significant evolutionary innovations in Earth's history, enabling the emergence of complex organisms and ecological relationships that transformed our planet's biosphere. While there has been progress in understanding this important biological transition that occurred millions of years ago, the exact processes that allowed single cells to aggregate into complex organisms composed by different cell types and differentiated tissues (as seen in most plants and animals), remain unclear. Becoming multicellular brings a series of new challenges, for example reduced nutrient accessibility to individual cells that can easily lead to cell death and the need for fundamental tasks to be divided across different cells. Using yeast as a model system, this project will investigate the genetic, molecular, and physiological steps involved in transitioning from existing as single cells into complex cell aggregates, that gave rise to the multitude of life forms present today. This research will provide crucial insights into the origin of complex life, as well as shed light into the processes involved in major biological transitions. This project will create a web portal that will facilitate the sharing of strains, data, and detailed methodological protocols that will facilitate the use of the yeast system by the broader community and the general public. It will also provide yeast evolution kits to high school and college classrooms, providing unique opportunities for students to witness and document evolution in action. The Multicellularity Long Term Evolution Experiment (MuLTEE) will provide a powerful approach to investigate the transition between single- and multi-celled life forms. It will evolve multicellularity in real-time through laboratory selection for increased organismal size in initially unicellular Saccharomyces cerevisiae yeast populations. Since the experiment's inception in 2018, these "snowflake yeast" have evolved from microscopic cell clusters into increasingly integrated organisms that are visible to the naked eye, developing remarkable innovations including entangled cellular architectures that distribute mechanical forces throughout the organism, differentiated cell types with specialized functions, and coordinated developmental programs. This grant supports the experiment's essential daily operations—particularly the daily transfers required for continued evolution, long-term cryopreservation of the full experiment, and collection of core genomic and phenotypic data. 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 Faculty Early Career Development (CAREER) award supports research looking to develop of a new approach to multifidelity scientific machine learning that combines data from both high- and low-fidelity simulations in a mathematically rigorous way, yielding new machine-learned models that issue high-accuracy predictions at low computational cost. In engineering design, predictive computational simulations enable design engineers to analyze the expected performance and cost of designs without needing to go through the expensive process of building and experimenting on physical prototypes. In typical settings, engineers have access to both high-fidelity simulations, which issue the most accurate predictions but at high computational cost, and low-fidelity simulations, which issue lower accuracy predictions more cheaply. If only high-fidelity simulations are used, the high computational cost of each simulation can limit the number of designs that can be considered, leading to sub-optimal designs. On the other hand, if only low-fidelity simulations are used, the low accuracy of the predictions can lead to less reliable designs. This CAREER award will support research focused on enabling engineers to create more optimal and robust designs across all engineering disciplines, ranging from space missions to biomedical devices to renewable energy systems, thereby advancing national defense, welfare, and prosperity. This award will also support the development of new educational modules for training both undergraduate engineering students and practicing engineers in industry in modern computational design methods, thereby promoting the development of a globally competitive STEM workforce. The intellectual contributions intend to yield new machine learning methods that enable design engineers to rapidly explore high-dimensional design spaces and quantify design-relevant uncertainties, opening up a new class of design problems that can be solved with the new methods. The multifidelity scientific machine learning approach intends to train models on both limited high-fidelity data and more abundant low-fidelity data by combining these data in a multifidelity control variate framework. Multifidelity control variates, which exploit correlations between high- and low-fidelity data, have been successfully used for uncertainty quantification for engineering design to yield provably more accurate uncertainty estimates at lower cost. The research activities of this project are to (i) develop and analyze new multifidelity control variate frameworks for linear and nonlinear regression, and (ii) validate the multifidelity learned models on a range of computational design problems including reliability analysis, optimization, and sensitivity analysis. The educational activities will (i) develop a short course on the methods for industry engineers, (ii) create curricular modules on computational design that complement existing undergraduate design courses, and (iii) involve undergraduate students and industry collaborators in the evaluation and validation of the research methods. These educational contributions will be widely disseminated at national conferences, promoting the development of modern computational design skills for all learners. 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

Mechanical metamaterials (MMs), with their unique structural configurations rather than their chemical compositions, offer a wide range of unprecedented mechanical properties, from enhanced toughness and energy absorption to advanced vibration damping and soundproofing capabilities. Recent advancements in additive manufacturing have enabled the creation of these intricate geometries, leading to materials that excel in diverse applications, including safety gear, aerospace, noise-canceling technologies, and impact-resistant devices. Designing MMs today involves a complex, time-consuming iterative process where experts intuitively define designs, validate them with physics-based simulations, and refine them through trial and error. Recent advances in generative AI and machine learning offer the potential to disrupt this design cycle by automating and accelerating the process. This research proposes the development of a computational pipeline that integrates AI-driven optimization techniques with advanced simulations to streamline MM design. Using a graph-based representation of MM structures, the system effectively reduces the design space of general 3D MMs from hundreds of thousands of dimensions to a compact space with 10–100 dimensions. The system will incorporate advanced optimization techniques that learn efficient design solutions and enable the transfer of insights across different material properties, for example, from stiffness to shear strength or impact resistance. Furthermore, the approach facilitates a more comprehensive understanding of material performance, enabling designs that balance multiple desired properties. Tested primarily on MMs, the approach will also be applicable to a broad range of scientific and engineering problems requiring the design of complex, high-performance materials. If successful, this project will establish a new class of AI-driven design tools that enable faster, more flexible, and sustainable material development. The project will also provide open-source tools, encouraging further research and collaboration in the field. The educational goal of this project is to create tools that inspire K–12 students to consider STEM pathways and to disseminate new curricula adapted to interdisciplinary 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 Awards · FY 2025 · 2025-06

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 Awards · FY 2025 · 2025-06

This project aims to revolutionize data analysis by harnessing the unique capabilities of immersive technologies, specifically virtual reality (VR) and augmented reality (AR), to design and develop more intuitive and powerful user interfaces. With the increasing complexity and volume of contemporary datasets, there is a pressing need for innovative interfaces that augment analytical capabilities, minimize cognitive load, and facilitate the rapid generation of valuable insights. VR/AR technologies represent the next generation of display and interaction platforms, transcending the limitations of physical screens by enabling users to experience information in two or three dimensions, at any scale, and in any physical location. The gaming and entertainment sectors have experienced considerable success as a result of substantial investments in VR/AR technologies. By integrating immersive technologies into data-intensive workflows, this research aims to reshape the future of workspaces and enhance human analytical capabilities in our increasingly data-driven world. Potential benefits include improved decision-making across sectors like healthcare, finance, and scientific research, while also democratizing data analysis for non-technical users. This could cultivate a more data-literate workforce and society, better equipped to address complex challenges through data-driven approaches. The project's technical approach is structured around three key aims: (1) Designing intuitive 3D data organization interactions that optimize explicit user input while supporting intelligent, implicit system assistance. (2) Developing embodied interactions for data analysis tasks, encompassing visualization authoring and data modeling, with integrated menu systems for precise configurations; and (3) Enabling an end-to-end immersive data analysis pipeline that supports seamless task transitions and enhances data provenance tracking. The research methodology integrates a multi-faceted approach, encompassing controlled laboratory studies, elicitation studies, and longitudinal evaluations, to facilitate the iterative design and validation of the proposed techniques. Through collaborations with domain experts from industry and national laboratories, we will ensure the real-world applicability of our research. This project will produce open-source tools that foster research and education in immersive analytics, thereby enhancing human analytical capabilities through spatial and embodied interactions in immersive environments. 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 I-Corps project is based on the development of a diagnostic test for preeclampsia, a pregnancy complication associated with high blood pressure. Preeclampsia is one of the leading causes of maternal morbidity and is estimated to occur in up to 10% of all pregnancies. When left untreated, preeclampsia can result in serious and sometimes fatal complications including stroke, seizure, kidney disease, heart disease, and preterm birth. Diagnosis of preeclampsia remains a significant challenge. The current diagnostic standard for preeclampsia relies on insensitive tests and outdated standards, allowing diagnosis only after evidence of organ damage in the mother. Also, preeclampsia tests generally need to be performed in clinical labs by trained personnel, leading to increased costs, decreased accessibility, and decreased likelihood of catching preeclampsia before it becomes dangerous to mother and fetus. This technology addresses these challenges by providing a preeclampsia diagnostic that can be used at home by untrained personnel with just a finger-stick volume of blood that is no more difficult to interpret than an at-home COVID test. This technology may improve detection, treatment, and management of preeclampsia and improve the outcomes of pregnant individuals with this condition. This I-Corps project utilizes experiential learning coupled with a first-hand investigation of the industry ecosystem to assess the translation potential of a preeclampsia diagnostic that can be used at home by untrained personnel with just a finger-stick volume of blood. The technology is based on a recently developed cell-free synthetic biology technique to measure proteins harnessed from nature and re-engineered to detect recently discovered biomarkers for preeclampsia in a mother’s blood sample. A visually interpretable color change that provides semi-quantitative screening for preeclampsia has been created that is similar to an at-home COVID test. The use of cell-free systems, which consist of purified bacterial extracts, enables easy reprogramming to create diagnostics for new diseases without the extensive, expensive, trial-and-error development cycle associated with existing point-of-care protein measurement techniques. This system is compatible with the lyophilization that is needed to enable shelf-stable products that can be distributed commercially. This technology may allow the development of at-home screening and diagnostic testing for preeclampsia and for many other different types of diseases in the future with the goal of improving patient outcomes. 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

A promising approach to decreasing the atmospheric burden of the greenhouse gas carbon dioxide (CO2) is to capture it from combustion sources and convert it to higher-value fuels and chemicals by using electrochemical cells powered by electricity generated from sustainable sources such as solar or wind energy. Catalysts are a key component of electrochemical reaction cells. The project combines theoretical, machine learning (ML) and experimental approaches to obtain fundamental knowledge critical to the design of highly efficient CO2 conversion catalysts. In addition, the project will provide early research opportunities for high school and undergraduate students, especially women and other underrepresented groups. A better understanding of chemical scaling relations and how to break them is urgently needed to achieve energy-efficient catalytic production of fuels and chemicals from CO2. However, understanding scaling relations and structure-property relationships under realistic operating conditions remains challenging. The project will utilize structurally precise catalysts as a testing platform and perform atomistic-scale simulations to reveal structures and catalytic properties under realistic reaction conditions. The first thrust will focus on catalytic scaling relations at the atomic level by developing and applying computational approaches (coupled with experimental validation) integrating quantum Monte Carlo, density functional theory, and molecular dynamics. Given the large chemical and structural space of these catalysts, the second thrust will couple fundamental insights with advanced machine learning techniques, including graph neural network models and interpretable machine learning methods, to efficiently design novel catalysts that break the identified scaling relations. The combined approach will be applied to structurally precise and atomically tunable catalysts (i.e., metal-nitrogen-doped carbon, atomically precise metal nanoclusters, and transition metal dichalcogenides) to explore the following three avenues for improving CO2 electrocatalytic reduction performance: I. Lowering the overpotential of the CO2 reduction reaction (CO2RR) through breaking the scaling relations between *CO and *CHO binding energies; II. Promoting C2+ products through breaking the scaling relations between *CO binding energy and C-C coupling barrier; and III. Suppressing the hydrogen evolution side reaction through breaking the scaling relations between *CO and *H binding energies. The project will deliver: (1) experimentally verifiable structure-activity relationships under operating conditions at the atomic level, (2) predictive computational methods and machine learning models for the study of electrocatalysis, and (3) highly active CO2RR catalysts with tunable selectivity. The project integrates research and education via workshops and training where discovery stimulates learning. Workshops will be hosted for grade 6-12 teachers, and relevant course modules will be developed for undergraduate students at Georgia Tech. 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 award provides partial travel support for U.S.-based researchers to participate in the international conference "Quantum Topology and Hyperbolic Geometry", to be held in Phu Quoc, Vietnam, from June 2–6, 2025, and the associated summer school at the Vietnam Institute for Advanced Study in Mathematics in Hanoi, from June 9–13, 2025. The conference will focus on various aspects of quantum topology, its connections to hyperbolic geometry and other fields in mathematics and theoretical physics. It will gather leading experts, young mathematicians, and graduate students from around the world, offering them a platform to share ideas and to collaborate. The summer school, designed for young mathematicians, will feature five minicourses and brainstorming sessions led by world-renowned experts, contributing to the development and training of new mathematical talent. Significant breakthroughs over the past few decades-such as the discovery of the Jones polynomial, the development of topological quantum field theories, and the introduction of gauge theory invariants, have revolutionized knot theory and three-manifold topology. These developments have forged deep connections between topology and diverse areas including number theory, Lie theory, and statistical physics, utilizing methods that extend beyond traditional algebraic topology. Quantum topology, in particular, has emerged as a pivotal field with substantial applications in biology, physics, combinatorics, algebra, and quantum computation. This conference seeks to build upon these advancements by exploring cutting-edge research in quantum topology and its interplay with hyperbolic geometry and other mathematical disciplines. By addressing a wide range of topics, the conference will promote interdisciplinary collaboration and stimulate innovative approaches that may lead to significant breakthroughs. Additionally, by engaging students and early-career researchers in key problems and frontier research areas, especially at the confluence of quantum and classical topology in three dimensions—the event will enhance their education and contribute to the vitality of the mathematical community. More information about the two events can be found at these websites: https://viasm.edu.vn/hdkh/School-Topo-Geo-2025, https://viasm.edu.vn/hdkh/Conf-Topo-Geo-2025. 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 I-Corps project focuses on the development of a wireless wearable technology system that enhances worker safety, ergonomic monitoring, and productivity in high-risk industries such as construction, manufacturing, and healthcare. The system addresses critical gaps in traditional workplace safety methods, which often rely on periodic audits and manual reporting. By providing workers’ real-time positions and motion analysis, the solution enables proactive intervention to reduce workplace accidents and associated costs. This technology has the potential to improve worker well-being, increase operational efficiency, and reduce downtime due to injuries. The project advances national priorities by promoting technological innovation in workforce safety and real-time monitoring, contributing to the progress of science and improvements in health, economic productivity, and workplace well-being. This I-Corps project utilizes experiential learning coupled with a first-hand investigation of the industry ecosystem to assess the translation potential of the technology. This solution is based on the development of an integrated system that combines wearable devices equipped with motion sensors and a cloud-based platform powered by artificial intelligence algorithms. The wearable devices continuously capture and monitor worker movements, enabling the recognition and classification of specific work activities in real time. The cloud platform aggregates and analyzes this data to provide actionable insights into safety, ergonomic assessments, and automated productivity tracking. Scientific advances include the application of machine learning models to accurately recognize worker activities and assess ergonomic risks, representing a significant improvement over traditional manual observation methods. Users benefit from proactive risk mitigation, enhanced worker safety, improved task efficiency, and the scalability of real-time operational monitoring across a variety of high-risk workplaces. 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

Scientific discovery and its successful commercial translation are critical components of U.S. global competitiveness. With growing international competition and a diminishing share of federal funding for research and development (R&D) compared to other sources, there is a need to better understand the effectiveness of government R&D funding programs. The Small Business Administration’s (SBA) Small Business Innovation Research and Small Business Technology Transfer programs (hereafter, SBIR/STTR), which provide non-dilutive funding to small U.S. businesses, are the most well-known of such programs. A long literature evaluates their effectiveness, and the U.S. Congress requires periodic review for reauthorization, yet critical elements of program design have been left out of quantitative evaluations and recipients acknowledge several potential design flaws. One overlooked element is the role of timing between SBIR/STTR grant rounds and specifically the impact of long waits between SBIR/STTR funding tranches. Advancing understanding around the role of timing could significantly improve program success and technology outcomes, furthering national technology and SBA priorities. This research evaluates the impact of accelerated government R&D funding for small technology business. The research uses data on federal R&D funding awards, data on commercialization, finance, and procurement outcomes, empirical modeling, and interviews. The project identifies policy levers for enhancing public programs that finance innovation, and explores the comparative advantage of such programs for some scientists and firms versus others, with particular attention to those innovators who have been overlooked and under-funded. The researchers examine 1) impacts of gaps in waiting time between government R&D funding tranches 2) the impacts on key firms that have been identified as priorities for support. The project has the potential to improve technology development and the ability of new, small technology firms to survive. It will provide actionable, evidence-based policy recommendations and have direct implications for federal programs, related state programs, and nonprofits that support firms in applying for SBIR/STTR awards. 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.