Georgia Tech Research Corporation

NSF Awards · FY 2025 · 2025-05

Ninety-five percent of deaf children in the US are born to hearing parents, most of whom never learn sign language well enough to teach it to their children. This puts these children at risk for delayed communication development compared to hearing children or deaf children of deaf parents. These communication delays may lead to social isolation, mental health issues, and lowered quality of life. Using educational games to help bridge these gaps has been proposed, but little educational software is designed for American Sign Language (ASL) and even less for sign languages of other countries. This project's goal is to create a framework for mobile device developers to use on-device sign language recognition to develop educational software for Deaf children and their hearing parents. The plan is to develop educational software, tools, datasets and machine learning models and release them via open-source so the research can be replicated for other sign languages world-wide. To establish that the tools can generalize across languages, the project team will develop versions of the software for both ASL and for Indian Sign Language (ISL). ISL is both less studied than ASL and has a large user population, making it an excellent second language to work in. To ensure the work is well-suited to the needs of deaf people, the work will be done in close collaboration with the Deaf community, and with participation of Deaf student researchers. The main technical focus of the project is to develop a framework that allows the rapid development of educational software that (a) can target multiple sign languages and (b) can be practically deployed without specialized hardware or setup requirements. The concrete educational software to be developed is based on two educational games for sign language previously developed by the research team that use word- and phrase-level ASL recognition in their gameplay. One key part of the project is the creation of robust datasets of ASL and ISL use, including fingerspelling, word, and phrase-level expressions. These datasets will both drive the creation of new recognition models and support linguistic analysis of ASL use. A second key element is designing novel algorithms for both recognizing sign language that are accurate enough to be useful and cope with disfluencies, flexible enough to support both adult and children's signing, and efficient enough to run on tablets and smartphones. The third key element is developing game design principles for educational software for sign language: avoiding the need for English-based instruction, appropriately responding to both well-formed and disfluent signing, and ways to assess and improve both users' receptivity and expressiveness around sign language. The research team will assess the effectiveness of both the software itself, in terms of its effects on children's language development, and the framework, in terms of its ability to be useful to other languages, developers, and 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 2025 · 2025-05

Non-Technical Abstract: The discovery of van der Waals (vdW) magnets opens a new pathway for designing innovative two-dimensional (2D) electronics to revolutionize the current information technologies. Taking advantage of their exotic material properties, vdW magnets serve as a leading candidate for building ultracompact, all-2D spin logic devices with tailored functionalities to outperform their conventional counterparts in device compatibility, stability, density, and tunability. Despite the enormous scientific expectation, the emerging 2D vdW spintronics research remains in its infancy in the current state of the art. One of the major challenges results from the very limited experimental tools capable of evaluating local material properties of vdW magnets at the nanoscale. In this project, the principal investigator plans to introduce scanning quantum sensing microscopy to timely address this grand challenge. An important goal here is to demonstrate the application of scanning quantum sensing techniques in a real 2D material environment to extract previously inaccessible information on vdW spintronic devices, providing valuable insights for future improvement of 2D spintronic technologies for practical applications. In parallel, this project will dedicate a major effort in increasing society’s awareness of some of the most exciting developments and challenges in spintronics, quantum sensing, and material sciences studies. The project will promote the participation of diverse groups of students at the forefront of scientific research. Proposed outreach activities include lectures, learning and demo materials for nearby technical colleges and high schools in the Atlanta metropolitan area, so that contemporary scientific knowledge can reach out to a significant amount of audience. Technical Abstract: vdW magnets with tunable lattice interactions and exotic spin properties are integral to cutting-edge scientific research, modern technologies, and therefore to a wide range of emerging applications. Taking advantage of the atomically thin nature, engineerable interfacial conditions, convenient material co-integration strategy, and readily established magnetic vdW proximity in artificial 2D stacks, vdW magnets provide an excellent platform to investigate exciting new physics and device merits that are unavailable in the 3D world. Novel functionalities are further added thanks to the emergence of room-temperature 2D magnetism, which significantly promotes vdW spintronic devices for practical applications. Here the principal investigator proposes to utilize nitrogen-vacancy (NV) centers to perform scanning quantum sensing of unconventional spin behaviors in room-temperature 2D magnet Fe3GaTe2-based vdW stacking devices. Exploiting the unprecedented spatial and field sensitivity, this project aims to directly visualize “field-free” deterministic magnetization switching and magnetic skyrmions in Fe3GaTe2, advancing the conceptual design and experimental development of vdW spin memory devices. It is further proposed to develop hybrid systems consisting of NV centers and functional vdW magnets to address the technical challenges faced by NV centers for practical quantum information applications. The proposed research will facilitate the development of 2D spintronic devices for emerging information communication, memory, and storage. By developing scanning NV quantum microscopy techniques and demonstrating their operations for cutting-edge material and device systems, this project proposes to provide a multimodal sensing platform, which can be readily extended to a large family of low-dimensional quantum materials, providing a new pathway for evaluating microscopic spin behaviors and device performance of innovative spintronic circuits. 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 · 2025-05

This grant provides travel support for Principal Investigators (PIs) with current active awards from the NSF Mechanics of Materials and Structures (MoMS) program to attend the 2025 MoMS Grantee Meeting, an integral part of the Society of Engineering Science (SES) Annual Technical Meeting. Scheduled for 12-15 October 2025, in Atlanta, Georgia, the meeting will serve as a platform to bring together NSF MoMS grantees, offering opportunities to strengthen collaborations and facilitate the exchange of cutting-edge ideas and research findings. The meeting will feature a series of poster sessions where grantees can present their research, encouraging peer review, feedback, and discussion. Panel discussions, led by invited experts, will dive deep into current challenges and opportunities in the field of mechanics of materials and structures. Topics will include advances in structural and soft materials, new insights into metamaterials, and the use of data-driven approaches to tackle complex engineering problems. The meeting will also include various social activities designed to promote networking and foster potential collaborations across disciplines and institutions. SES has long been a catalyst for advancing research in mechanics, particularly in the areas of structural materials, soft materials, and metamaterials. By providing a forum for the exchange of ideas between seasoned researchers and emerging talent, the meeting will help drive forward-thinking discussions and innovative solutions to pressing challenges in the field. The event will also offer invaluable mentorship opportunities, where established PIs can share insights on how to navigate the research landscape, secure NSF funding, and develop competitive proposals. The 2025 MoMS Grantee Meeting aims to further solidify the mechanics of materials and structures community by strengthening its connections and providing the next generation of researchers with the tools and guidance needed to thrive in their academic and professional careers. This gathering will foster an environment of collaboration and innovation, propelling forward research that is critical to advancing the mechanics of materials and structures, as well as addressing emerging global challenges. 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 · 2025-05

Dissolved oxygen is essential for supporting life in marine habitats and for controlling the cycling of carbon and nutrients in the global ocean. Historical observations have shown that oxygen concentrations are declining in many parts of the oceans, termed “ocean deoxygenation”. The causes of the decline are linked to changes in surface water temperature and its impact on oxygen solubility, along with variations in the strength of the biological pump, ocean circulation, vertical mixing, ventilation, and biochemical processes. It is difficult to calculate exactly how much ocean oxygen has been lost to the atmosphere and how much has been redistributed within the ocean interior. This project will compare observation-based gridded oxygen datasets with the purpose of generating comparable four-dimensional (space and time) estimates of global oxygen distribution, assessing their similarities and differences and promoting scientific understanding of the drivers of ocean deoxygenation. The project will promote workforce development through support of an early career investigator, a postdoctoral researcher, a graduate student, and several undergraduate interns. The project has two primary goals. The first goal is to identify the causes of disagreements between different gridded oxygen datasets by conducting an intercomparison of oxygen datasets from common in situ observational and model-based profiles. Many factors can affect estimates of a deoxygenation trend: existing observational studies use different sets of raw data, measurement platforms, data quality control metrics, land-ocean masks, vertical and horizontal grids and interpolation methods. Because variations in any of these factors can lead to different estimates, it is difficult to make direct comparisons and determine the causes of disagreements among datasets. Standardized protocols will be applied to isolate the interpolation method as the only source of discrepancy and to assess uncertainties in global oxygen inventory trends. The second goal is to test hypotheses for the underlying causes of ocean deoxygenation. The suite of new datasets will be used to evaluate the roles of oxygen solubility, the biological pump, and physical and biogeochemical processes driving global ocean deoxygenation. Further, the mechanisms driving the expansion of the tropical ocean oxygen minimum zones will be explored. The use of novel datasets with unprecedented spatio-temporal resolution in these analyses will enable new insights into global and regional oxygen content changes. The suite of gridded oxygen datasets will be made available via public data repositories, and the results of the product intercomparisons and deoxygenation analyses will be disseminated via open-access papers. Collaboration with the Scientific Committee on Oceanic Research (SCOR) Working Group 168 will enhance access and utility of gridded oxygen datasets through sharing and exchange of ideas, experimental protocols and data products. 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 · 2025-05

This award provides support to a CBMS conference on Legendrian links and the microlocal theory of sheaves. This will be a week-long workshop to be held June 9-16, 2025 at the Georgia Institute of Technology. The workshop will be centered around a mini-course consisting of ten lectures by Dr. Roger Casals and will be supplemented with several talks by other experts. There will also be sessions where participants can engage with the material by working on problems and discussing the lectures with experts. While the organizers hope to have broad participation from across the country, they will focus on participants from the Southeast. There are currently many universities in the region with active research groups in symplectic and contact geometry, such as the University of Georgia, Duke University, the Georgia Institute of Technology, the University of Alabama, and Louisiana State University, but few have experts in the topic of this workshop. So this will be a valuable opportunity to expand the tools available to researchers in the Southeast. This workshop should be of interest to a wide range of graduate students and researchers in nearby fields. In more detail, workshop will focus on recent developments in the study of Legendrian links in contact 3-manifolds. Specifically, the newly discovered connection between Lagrangian fillings and cluster algebras, as well as its method of proof, using the microlocal theory of sheaves. This is a topic of current research interest and there has been significant activity in the area, such as establishing that the sheaf moduli associated with any Legendrian (-1)-closure of a positive braid admits a cluster structure. Several open questions have now been solved based on this result and the study of weaves, including the existence of infinitely many Lagrangian fillings, the existence of cluster structures on Richardson varieties, and the Fock-Goncharov duality conjecture for braid varieties, among others. Several generalizations and related results are also being studied, including a conjectural classification of Lagrangian fillings. The conference website is https://sites.gatech.edu/ttss2025/. 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

Lipids and proteins assemble to carry out a range of biological functions. Their assemblies are a universal feature of nearly all cellular pathways, but the mechanisms that drive their interactions have historically been difficult to map and understand. The project will study how lipids and protein molecules assemble to carry out their function via the development of new computational tools to model all types of lipid/protein assemblies, and the in depth study of the mechanism by which lipids interact with different families of proteins that are important for immune system function. This project also aims to design and implement a course-based research experience focused on contemporary biophysical and biochemistry tools to characterize lipid/protein interactions where the research topic, approach, and inclusion of peer mentoring between graduate and undergraduate students drives pedological innovation. This project will the public to improve understanding of science and technology innovations in modern research. This project seeks to delineate the wide range of mechanisms by which the eight classes of lipids engage with their protein partners, and to understand how lipid/protein assemblies are regulated at the molecular level. The first objective is to develop new machine learning-based predictive tools to model specificity and atomic structures of lipid/protein complexes. Further objectives apply integrated computational biology, biophysical assays, and structural biology approaches to (i) define the structural basis by which lipid transfer proteins participate in lipid antigen selection and editing on CD1 immunoreceptors, which underpins immune recognition, and (ii) to characterize how unique repertoires of lipid antigens are selected by the coordinated effort of several lipid transfer proteins. The research is expected to provide the mechanistic basis for specificity and selectivity of lipid/protein complexes relative to peptide/protein complexes and further define the roles of lipid binding proteins in biology. This support also establishes educational and outreach programs to improve enrollment, engagement, community, and retention students in STEM 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 · 2025-04

People can easily be overwhelmed with data when making decisions, such as deciding which healthcare treatment is appropriate or which political candidate to vote for. When overwhelmed by data, people tend to seek and interpret information in a way that supports their preexisting beliefs. This phenomenon is often referred to as confirmation bias. In data communication and visual analytics, confirmation bias can be especially nefarious, even for experienced analysts. Although there is a misconception that statistical models and visualizations present objective truths, in reality, choices in the collection, handling, analysis, and presentation of data can bias people into overly relying on their pre-existing beliefs. This project will closely examine confirmation bias in data analysis by (1) creating models that show how existing beliefs and analytic goals can impact data-driven decision-making, and (2) designing novel analytic interfaces that help analysts make less biased decisions by intelligently suggesting evidence that may disprove a belief. The project team will also create educational materials and collect empirical datasets to help data analysts, researchers, and members of the public think about confirmation bias in visual data communication and interpretation. This project aims to increase understanding of how confirmation bias manifests in real-world visual data analysis tasks and to develop and evaluate bias-mitigation interventions. The project is structured around four research thrusts. As measuring confirmation bias requires capturing an individual’s beliefs, the researchers will first investigate experimental methods to accurately capture a person's beliefs and mental representations about data patterns and trends in an analytic setting (Thrust I). The researchers will then leverage these methodological findings to measure and model the effect of confirmation bias in low-level visual analytic tasks such as finding correlations (Thrust II), then examine higher-level compositions of these tasks in more realistic analysis settings (Thrust III). Finally, the researchers will design and develop four bias-mitigation interventions to be incorporated into real-world visual analytic tools such as Data Voyager and Jupyter Notebooks, recruiting professional analysts to evaluate them in digital field studies (Thrust IV). This research agenda will advance the understanding of confirmation bias and provide promising interventions to empower data analysts to make better decisions. The researchers will also develop coursework and initiatives that bring together computer science, psychology, and ethics to advance practice and education around visual data analytics. 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

The IEEE International Conference on RFID (IEEE RFID) is the premiere venue to present multi-disciplinary cutting-edge research in RFID, wireless sensor networks, energy-harvesting, localization, and low-energy wireless communications -- topics which all touch on next-generation wireless technologies. These cutting-edge research areas are key enablers for the Internet of Things (IoT) and machine-to-machine communications. This travel grant will allow continued growth and participation of early-career U.S. researchers in this strategically important community. With a history of 19 years, the IEEE RFID conference series has become the major event in this community with strong emphasis on industry-academic partnership, attracting more than half of the attendees from industry. The 19th IEEE RFID conference will be held in Atlanta, GA on April 22-24, 2025 and co-located with two other IEEE conferences, the IEEE Digital Twins and Parallel Intelligence (DTPI) and the newly launched IEEE Additively Manufactured Electronics Systems (AMES), as well as a number of networking events and industry sessions that are part of the Cyber-Physical Radio Week. To support the training of the next-generation scientists and engineers working on wireless technologies and applications, this NSF travel grant will provide travel support for 20 U.S. students or post-doctoral researchers to attend and present their papers at the conference. Their participation in the conference and co-located events will be valuable opportunities for them to learn the newest research and development trends in this field and interact with other attendees including industry participants. This travel grant will contribute to the U.S. STEM workforce development and have long-term and broad impacts on the career developments of participating students and postdoctoral 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 2025 · 2025-04

This I-Corps project focuses on the development of an artificial intelligence (AI) driven technology that predicts protein dynamics, addressing a major challenge in drug discovery. Traditional methods for modeling protein motion require months of computational time and access to supercomputers, making them expensive and impractical for widespread use. Understanding protein dynamics is crucial for identifying new drug targets, optimizing lead compounds, and reducing failure rates in early-stage drug development. By improving the accuracy and speed of drug discovery, this technology has the potential to accelerate the development of new treatments for diseases that currently lack effective therapies. The potential commercial impact is substantial, as pharmaceutical companies and research institutions require better tools to streamline the drug development process, ultimately leading to more efficient pipelines and lower costs. This innovation may also expand the target space by identifying hidden binding pockets, opening new opportunities for therapeutic interventions. 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 a deep-learning model that predicts dynamic protein conformations using physics-based force fields. Unlike existing approaches that rely on extensive datasets and prolonged simulations, this project introduces an active learning framework that enables efficient, on-the-fly model training. This approach is further extended through novel methodologies that generate new protein conformations optimized for virtual screening and assess ligand-induced conformational changes, including binding free energy calculations. By integrating diffusion-based deep learning models with molecular physics, this project advances computational biology by providing a scalable, adaptable tool for drug discovery. The insights gained from this project will help refine the technology's commercialization strategy and establish its role in the evolving landscape of artificial intelligence driven drug discovery. 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

NON-TECHNICAL DESCRIPTION Organic semiconductors are promising materials for flexible electronics such as solar cells, displays, and sensors. For many applications, their electrical conductivity must be increased by adding positive (p-type) or negative (n-type) charges. This process is known as molecular doping. While there are many p-type organic semiconductors, n-type organic semiconductors are far rarer. Moreover, they are less stable in ambient conditions, and it can be difficult to control their electrical conductivity. Advancing the field of organic electronics will require development of new electronic materials and processes for n-type doping. The goal of this project is to enhance the efficiency of n-type doping using a catalyst, a compound that enhances the speed of a chemical reaction. The investigator will combine synthetic strategies for new organic semiconductors with selection of n-type dopant-catalyst pairs to optimize the doping process. This research will result in a new class of n-type organic semiconductors with high-performance and enhanced stability. This research will also promote understanding on how the new n-type materials affect device performance. The PI’s educational goal is to foster a positive perception of organic electronic materials and devices. This will be accomplished through outreach to middle and high school students, coupling course instruction to undergraduate and graduate research projects, and providing internship opportunities at a start-up company to foster entrepreneurship. TECHNICAL DESCRIPTION The doping of inorganic materials has been instrumental in the progress of the semiconductor industry and advances in numerous fields including medical diagnostics, environmental science, and homeland security. For organic semiconductors, several strategies have yielded doped π-electron solids with greatly enhanced optical and electronic properties as well as novel materials, physical phenomena, and device concepts. However, these advances were largely enabled by p-doped (hole transporting) materials, while useful n-doped (electron-transporting) materials have been limited in chemical accessibility, doping efficiency, and environmental stability. This limits their use in devices where both p-type and n-type semiconductors are required. Recently, the PI and coworkers discovered that metal nanoparticles (e.g., Au) can catalytically assist/accelerate the doping of representative organic semiconductors by molecular dopants. This research project will: 1) Elucidate the mechanism of the catalytic n-doping process and expand the scope to other type of catalysts. 2) Explore other dopants and semiconductors, implementing to date unexplored structural design and synthetic strategies. 3) Characterize catalyzed vs uncatalyzed doped film properties in terms of composition, electronic structure, morphology, microstructure, and opto-electronic response. Thus, this study plans to bring n-doped organics to unprecedented levels in terms of molecular/macromolecular materials designs, their accessibility in scale, and their useful opto-electronic properties. 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

Integrable systems are a fundamental mathematical tool in multiple subfields of mathematics and physics, and more importantly, are instrumental in building bridges between seemingly disparate areas. This project deals with the recent resurgence of integrable systems in the context of algebraic geometry and representation theory, motivated by the study of quantum field theories. The results of the research are expected to elucidate the phenomenon of three-dimensional mirror symmetry originally discovered in theoretical physics. Broader impacts include establishing an interdisciplinary program for mathematics and physics students. The project contains undergraduate and graduate student research topics and aims for vertical integration of research and education. The PI will also co-organize several conferences designed for early-career researchers. The project is jointly funded by the Geometric Analysis program, the Algebra and Number Theory program, and the Established Program to Stimulate Competitive Research (EPSCoR). The project deals with the geometric realization of several points of view on the Bethe ansatz approach to quantum integrable systems. One approach uses quantum Knizhnik-Zamolodchikov equations emerging naturally in the enumerative geometry of Nakajima quiver varieties. Another one relies on the study of QQ-systems, generalizing the relations satisfied by the Baxter operators. The geometric realization of QQ-systems is derived from the properties of the difference equation version of oper connections on the projective line, generalizing the correspondence between oper connections and Bethe equations for the Gaudin model. The PI plans to show how these geometric points of view on the Bethe ansatz are bonded together within the framework of the quantum q-Langlands correspondence. This is expected to lead to new results in the study of the mathematical formulation of three-dimensional mirror symmetry. 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 ubiquity of cameras and other sensors in our environment coupled with advances in computer vision and machine learning has enabled several novel situation awareness applications combining sensing, processing, and actuation. Many of these applications are latency sensitive, network bandwidth hungry, and geo-distributed. Edge computing has emerged as a new trend in catering to these computational needs by moving computers physically near the sensors, and is driven in part by low-cost processing resources such as Raspberry Pi and NVIDIA Jetson Nano. In the 5G/6G evolution, cell phone companies are moving network-level packet processing to geo-distributed edge data centers to amortize the computational cost of the proliferation of the wirelessly connected devices. Function-as-a-Service (FaaS) (which allows users to run code without managing server infrastructure) is an appropriate allocation paradigm to increase the utilization of resource-constrained Edge sites that host such applications, but to date its design has targeted Cloud datacenters. The project’s novelty rests in addressing the gap that exists in efficient system support for FaaS for a geo-distributed Edge ecosystem. The project’s broader significance and importance rest on the assertion that Edge computing could well be the next wave of disruption due to the emerging influence of AI and ML pervading all walks of future human endeavor. Thus, the technology nuggets from this project could spur such disruption in the technology landscape. The project is an end-to-end software system architecture for latency-sensitive situation awareness applications using the FaaS paradigm on geo-distributed Edge infrastructure. Specifically, the project makes the following advances to the state-of-the-art: programming idioms and associated runtime machinery that ensure the spatial and temporal correctness of the applications deployed on the Edge infrastructure without requiring distributed systems expertise which is largely unavailable to most developers; software control plane techniques for efficient federated orchestration across the Edge infrastructure while preserving the application’s service level objectives (SLOs) and ensuring spatio-temporal application correctness; data plane techniques for the efficient management of application state at each Edge site with respect to spatial and temporal correctness; and nimble execution environments to quickly launch application components while conserving limited Edge resources. 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 · 2025-02

Advancements in artificial intelligence (AI) are transforming scientific discovery, enabling breakthroughs that were once thought unattainable. In biology, AI-driven models like AlphaFold have revolutionized protein structure prediction, achieving unprecedented accuracy in determining 3-dimensional structures from amino acid sequences. This capability has opened the door to applications ranging from drug discovery to protein engineering. Despite these successes, the inner workings of such models remain a black box, hindering their broader scientific utility and the development of fundamental theories. This project aims to illuminate the internal activity and their correspondence to bio-physical principles by developing advanced visualization and interpretability tools. These tools will allow researchers to gain unprecedented insights into how these AI models operate, fostering a deeper understanding of protein folding processes. By providing open-source software and an interactive Science Gateway, the project will democratize access to these capabilities, empowering a wider range of scientists to explore and refine protein folding theories and showcasing the role of AI in advancing fundamental scientific knowledge. VizFold tackles the critical challenge of understanding and interpreting the mechanisms underpinning protein folding predictions made by AI models like AlphaFold. While these models excel at accurately predicting protein structures, the processes and principles they use remain opaque, limiting their broader applicability and the development of scientific theories. To bridge this gap, the project focuses on three key objectives. First, it will develop visualization tools to examine activation propagation within the specialized architectural components, including attention mechanisms and structure modules unique to AlphaFold-like models. Second, it will enhance interpretability by applying probing techniques, dimensionality reduction and Layerwise Relevance Propagation, to elucidate how these models encode folding processes. Third, it will build an interactive Science Gateway powered by Cybershuttle, enabling researchers to visualize and analyze model outputs in real-time while offloading computational tasks to high-performance infrastructure. By achieving these goals, VizFold will advance computational biology, support non-expert users, and lay the foundation for a new physical-chemical understanding of protein folding, paving the way for future innovations in AI-driven science. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-02

With support from the Chemical Structure and Dynamics (CSD) program in the Division of Chemistry, Professors Cicerone and McDaniel of The Georgia Institute of Technology will develop a novel framework for rationalizing liquid dynamics by connecting fundamental molecular relaxation events to macroscopic properties. Their approach will combine experimental measurements of relaxation dynamics with simulations to elucidate how molecular excitations facilitate motion across the complex potential energy landscape in glasses and liquids, with intrinsic barriers correlated to thermodynamic quantities of the system. The team will use a combination of neutron scattering, Raman spectroscopy, and molecular dynamics with ab initio and machine-learning potentials to validate their framework and explore the universality of liquid behavior across several different molecular structures, classes of inter- and intramolecular forces, and pressures. Their studies could enable the prediction of macroscopic relaxation and transport processes of liquids based on fundamental thermodynamic parameters, which could lead to new design principles for tailoring bulk properties for various technical applications. The Cicerone and McDaniel groups consistently provide research experience and opportunities for undergraduate students and participate in several educational outreach activities each year, introducing scientific concepts to local K-12 students. Transport and relaxation in liquids and glasses occur through microscopic cooperative rearrangements on picosecond timescales and Angstrom length scales. The Cicerone and McDaniel labs have recently quantified the population of particles involved in these elemental relaxations and have shown that the activation free energy for these collective relaxation processes is proportional to that of crystallization for simple liquids. In this project, they will determine how to find these activation barriers for complex systems without a well-defined melting point. They will also connect molecular structure with the topology of sequential microscopic reorganization events that lead to structural relaxation. Finally, they will focus on systematically adapting ab initio force fields to reproduce THz collective motion observed in experiments. These studies will seek to quantitatively reproduce the temperature and pressure dependence of transport and dynamics measured using quasi-elastic neutron scattering, THz Raman scattering, and simulation. Success in these steps will help realize improved efficiency in designing materials and liquids with targeted properties. 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 · 2025-01

Online resource allocation lies at the core of computer science, operations research, and economics, addressing the persistent tradeoff between utilizing resources to seize immediate opportunities and conserving them for potentially better options in the future. Modern applications, such as online advertising, data center management, and ride-sharing operate on a massive scale, necessitating algorithms that learn from historical data, remain robust to outliers and corruptions, and adapt to diverse and dynamic scenarios. This project seeks to tackle these key technical challenges by developing innovative and effective solutions that will expand the boundaries of existing theory, address critical knowledge gaps, and pave the way for new innovations in online resource allocation. In addition to closely advising graduate students and developing curriculum, the educational components of this project include organizing a summer school on online algorithms, auction design, and data-driven algorithms, preparing a comprehensive textbook on Online Decision Making, and hosting a research workshop featuring invited talks from leading experts in academia and industry. The project focuses on three key directions in online resource allocation: (i) data-driven algorithms, which leverage limited historical data to infer problem structures and handle unknown input distributions; (ii) robust algorithms, designed to withstand adversarial corruptions, noise, and graphical correlations, ensuring resilience against errors in input models; and (iii) versatile algorithms, capable of addressing diverse objectives and constraints across a wide range of industrial applications. Tackling these directions involves addressing fundamental questions in economics, convex geometry, and optimization, including the role of prices in resource allocation, the structure of norms defined by general convex bodies, and the design of simple/fast algorithms for stochastic convex optimization. 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 · 2025-01

This project explores the gaps in knowledge, skills, and experiences that students may need to gain outside formal learning environments in computer science education and seeks to understand how these gaps impact students' success. By evaluating students' success as they navigate both formal (classroom) and informal curricula (e.g., makerspaces, internships, extracurricular clubs), we will develop learner-centered solutions to support their understanding of computing concepts and their gain of skills. The significance of this project lies in its potential to make CS education more comprehensive. In addition, this project addresses growing impact of artificial intelligence (AI) in education by examining the relationship of exposure to AI (e.g., large language models) outside of the class and student success in programming environments in the classroom. Our findings will benefit society by understanding and improving the educational experiences of all students and enhancing their success in computing programs. The three-year research program will investigate gaps in CS education through three primary strands: (1) identify factors from the formal and informal curricula, which when not available to students, could pose risks to students' mental health, such as anxiety, depression, and the impostor phenomenon; (2) study students' interactions with programming environments and large language models (LLMs) outside of class to characterize effective scaffolding strategies and address technical challenges in the classroom; and (3) evaluate the impact of makerspaces on students' creativity, exploratory skills, and sense of belonging. The project's methodology combines qualitative ethnographic methods, participatory design, and quantitative experimental and quasi-experimental approaches. This project will emphasize the experiences of all students in computing, aiming to create robust learning environments. The research will provide valuable insights and guidelines to improve CS education, ultimately reducing dropout rates and enabling students' success. 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 · 2025-01

This research project imagines the future of work in precision manufacturing where the spatial and causal reasoning and decision-making abilities of workers are augmented through teaming with intelligent extended reality (IXR) technologies. Evidence suggests that the newer wave of automation in manufacturing is not so much to replace workers but rather to complement human work to increase precision, safety, and product quality. Yet, U.S. manufacturers are not adequately addressing the changing nature of skill requirements which is anticipated to leave 2.4 million manufacturing jobs unfilled by 2030. This project will address the urgent need for breakthrough technologies that enable workplace-based learning and rapid upskilling of the manufacturing workforce on complex, cognitively demanding, and hard-to-automate tasks. The project will focus on precision machining and inspection in the aviation industry as the specific work context for building and validating the IXR technologies, which is also expected to inform the technology development in other industries such as medical, automotive, semiconductor, and defense. The convergent research team will create new technological pathways to enable intelligent worker-XR teaming and advance the fundamental understanding of its impacts on labor economy and worker learning and innovation. This project aims to create new perspectives, methods, and discoveries to unleash the full potential of America’s manufacturing workforce, and as such, strengthen national prosperity and economic competitiveness in precision manufacturing. This project brings together several disciplines, including engineering, learning sciences, social sciences, economics, computer sciences, psychology, and workforce development. The investigator team is structured to achieve multiple convergent goals across the three dimensions of the Future of Work at the Human-Technology Frontier: (1) The Future Work dimension will investigate the changes in employer skill requirements for precision manufacturing including education, years of experience, and actual skills, using a proprietary database of 160 million online job vacancies. Expert interviews, firm-level surveys, and in-depth case studies will investigate training and upskilling practices for incumbent and entry-level workers, explore accessibility of the IXR approach for certain groups of workers and firms, and identify economic barriers and opportunities for adopting the IXR technology. (2) The Future Technology dimension will advance the fundamental understanding of how new sources of multimodal data captured by XR devices, digital thread, IoT, and cloud-based analytics can be harnessed to interpret, predict, and guide the behavior of precision manufacturing workers. A novel IXR technology will be built and validated that adapts the scientific methods of computer vision, natural language understanding, and inference engines to provide intuitive and personalized assistance to workers performing complex reasoning and problem-solving tasks. (3) The Future Worker dimension will generate new knowledge about the affordances of worker-XR teaming to support the development of workers’ adaptive expertise for increasingly complex manufacturing tasks, building on research from the learning sciences that examines cognitive processes associated with complex reasoning and problem solving. It is expected that the knowledge generated in this project will elicit new pathways for the design of future collaborative human-technology systems for training adult workers beyond XR. This project has been funded by the Future of Work at the Human-Technology Frontier cross-directorate program to promote a deeper basic understanding of the interdependent human-technology partnership in work contexts by advancing the design of intelligent work technologies that operate in harmony with human workers. 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 · 2025-01

The broader impact/commercial potential of this I-Corps Hubs project is the development of a transformative entrepreneurial ecosystem that enhances the nation's economic and industrial competitiveness through fostering entrepreneurship and deep technology workforce development. Spanning four southern states, the Hub addresses regional inequities in access to capital, experienced entrepreneurs, mentors, and advisors for deep technology researchers. By leveraging collective strengths of partner institutions with extensive backgrounds in research, entrepreneurship, and innovation, the project aims to catalyze the ecosystem through mitigating technological, expertise, and capital disparities. The Hub will deploy a regional I-Corps infrastructure with full-time access to entrepreneurial training at scale, assisting researchers in applying customer discovery principles to technology translation. This project serves the national interest by promoting increased participation in national I-Corps programs throughout the region. The anticipated outcomes include a commercialization roadmap with enhanced academia-industry partnerships, and equitable access to opportunities across the South's innovation landscape. By focusing on increasing engagement with the Southern community in terms of startups and technology commercialization, the project seeks to increase interest in science, technology, engineering and math (STEM) fields. This I-Corps Hubs project is based on the development of a comprehensive regional innovation network that combines the strengths of varied academic institutions. The project's goal is to create a scalable model for translating research into commercial ventures with societal benefits, including opportunities for all entrepreneurs. The approach involves integrating best practices from existing I-Corps Hubs programs. Key methods include collaborative research, commercialization activities, and specialized training programs. The project will actively monitor success metrics for all Hub activities. This approach extends beyond traditional innovation support, incorporating research to understand the impact of differing environments on participation. The project aims to make significant contributions to entrepreneurship, innovation best practices, and regional economic development. The goal is to positively impact the region’s economy from the establishment of new businesses and commercialization of new products and services that have benefitted from the Hub’s I-Corps training. The imperative to expand the region's economic base serves as a strong motivator, potentially increasing interest in science, technology, engineering and math (STEM) fields and yielding access to opportunities across the South's innovation ecosystem. 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 · 2025-01

This project focuses on identifying and quantifying the atmospheric emissions from a chemical release that occurred as the result of a fire at the BioLab facility in Conyers, Georgia on September 29, 2024. The event resulted in a large-scale release of chlorine and bromine containing compounds forming a chemical plume that impacted Atlanta and the surrounding areas for several weeks and led to the evacuation of approximately 17,000 people and a stay-at-home order for the surrounding county. Biolab is a manufacturer that produces chlorine and bromine containing chemicals for pools and spas. The results from this project will increase the ability to predict the impacts of chemical incidents such as this one and enable a better understanding of how to address accidental chemical emissions in the future. The project includes the following tasks to: (1) finish calibration of the inorganic bromine data; (2) develop a lower limit for the observed HNCO and CH2N2 concentrations; (3) prepare a publication describing the Cl2, Br2, HNCO and CH2N2 data and archive the data to a publicly accessible site; (4) analyze mass spectra associated with the plumes and identify as many compounds as possible; (5) calibrate the species identified, prioritizing them based on toxicity; and (6) report on all species detected in the plumes in a second publication. This effort will inform local communities near the site who are concerned over the identity and level of chemicals to which they were exposed. A graduate research assistant will be trained and supported for 9 months during the project. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-01

Manufacturing, medical laboratory, construction, and many other jobs require workers to learn complex physical “psychomotor” tasks that combine both perceptual and motor skills. These are often taught using an apprenticeship model on real jobsites, which raises both productivity and safety risks for workers. Further, relatively little is known about how to assess trainees’ skill levels in these tasks and to adapt training practices based on those assessments. This project tackles these problems by developing a new generation of intelligent tutoring systems that combine extended reality (XR), artificial intelligence (AI) and Internet-of-things (IoT) technologies to support training and assessment of complex skills required by modern, highly automated manufacturing facilities. The high level idea is that new sources of data captured by XR headsets, wearable devices, cameras, and IoT sensors can be used to build models of psychomotor skill development and new methods for providing personalized, just-in-time coaching guidance. Through partnerships with manufacturing consulting firms, local community colleges, and K-12 schools, the project will enhance the skill development of a diverse population of learners and professionals and expand interest in advanced manufacturing careers. The project team brings together expertise in engineering, cognitive psychology, learning sciences, game design, and XR, to make fundamental contributions to both learning science and learning technologies around just-in-time, personalized, context-aware provision of learning scaffolds for manufacturing workers learning new skills. On the learning side, the project team will examine the stages of expertise development for specific psychomotor tasks, and the effectiveness of adaptive interventions on learners’ engagement, performance gains, and accuracy. A virtual reality (VR) game in an advanced manufacturing scenario will be used to collect ecologically valid baseline data and prepare more novice learners for real-world task performance. On the technology side, the project team will build and validate an intelligent XR tutoring system to accelerate the learning of psychomotor tasks with high complexity that arises from task structures and human information processing requirements. The innovative aspects of the technology include data-driven activity understanding (e.g., task step identification and error detection) and user modeling (e.g., cognitive load detection), through novel multimodal AI architectures designed to process and fuse data captured from augmented reality (AR) headsets, wearables that capture physiological data, cameras, IoT sensors, and manufacturing machines. Both learning and technology innovations will be validated through extensive laboratory studies; together, the work will lead to an intelligent feedback algorithm to dynamically adapt the nature, frequency, and depth of feedback to the expertise of the learner to facilitate optimal learning and speed-to-competence. 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 · 2025-01

One of the key challenges in advancing artificial intelligence is improving AI models' ability to generalize across different tasks, particularly when encountering data distribution shifts from training to real-world applications. Data distribution shifts are encountered when the statistics of the learned model are different from the real-world statistics. This project focuses on enhancing machine learning with graph-structured data, a direction with the potential to revolutionize scientific discoveries in particle physics and biochemistry. In addition to contributing to advancements in these scientific fields, the project also emphasizes knowledge dissemination through curated new datasets, scalable software solutions, workshops, and tutorials. Additionally, it encourages undergraduate and K-12 students from underrepresented groups to engage in research projects through the STEM program at Georgia Tech. This research will tackle the problem of distribution shifts in graph machine learning through two main research thrusts. The first thrust, graph structure calibration, aims to develop methods to estimate and mitigate shifts in entity connection patterns within graph-structured data from training to evaluation phases. It will address tasks such as node classification, regression, and link prediction. Additionally, this thrust will explore the accumulation of graph structure shifts and address challenges in open-world settings. The second thrust focuses on developing foundational models for graph data that are provably expressive, generalizable, and scalable, and investigating robust fine-tuning approaches for these models. The developed methodologies will be evaluated on particle representations in high-energy physics applications and molecule representation tasks. 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 · 2024-12

This project proposes to convene a workshop in February 2025 at the Georgia Institute of Technology Campus in Atlanta to gather early career researchers (ECRs) with an emphasis on supporting participation from Minority Serving Institutions (MSIs) and historically underrepresented populations to create a comprehensive needs assessment for training, cyberinfrastructure, and keys for open science adoption. The two-day workshop will be a collaboration between the South Big Data Hub (SBDH), CODATA-RDA Schools of Research Data Science (SoRDS) program, and the ScienceCore Heuristics for Open science Outcomes in Learning (SCHOOL) project (MacManus et al. 2023) of the NASA Transform to Open Science (TOPS) mission (Gentemann et al. 2021). The workshop will be hosted by SBDH in coordination with their Supporting Minority and Regional Training in Data & AI for Researchers of Tomorrow (SMART-DART) 2025 training program. This award is jointly funded by the NSF Office of Advanced Cyberinfrastructure and the Directorate for Social, Behavioral and Economic Sciences (SBE). 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 · 2024-12

The evolution of multicellularity from unicellular ancestors fundamentally altered the history of life, giving rise to an astounding diversity of complex organisms. While scientists have made significant progress understanding this transition in organisms that develop through cell division (like most plants and animals), a major question remains: how does multicellularity evolve in organisms that form groups by cells coming together and sticking to each other - a process called aggregation? Aggregative systems face a fundamental constraint: multicellular organisms can be comprised of unrelated cells, so natural selection is more likely to act on the traits of cells than the traits of the collective as a whole. Using yeast as a model system, this research investigates whether the ability to recognize and preferentially stick to genetically related cells (kin recognition) was crucial for the evolution of aggregative multicellularity. This project will provide unique insights into the origins of increasingly complex life and contribute to our understanding of the processes driving major evolutionary transitions. Additionally, this project will support broadening participation in STEM through mentorship of a female postdoctoral researcher from a low-income background, and will provide research experiences for undergraduate women. This project will test the hypothesis that kin recognition enables the evolution of multicellular adaptations by increasing the covariance between collective-level traits and the underlying genotype. The researchers will test this hypothesis using experimentally evolved strains of flocculating yeast, Saccharomyces cerevisiae, that have developed the ability to recognize and preferentially bind to related cells. Aim 1 will identify the molecular mechanisms of kin recognition through genetic engineering and competition experiments examining whether changes in the FLO1 binding protein alone or coordinated changes between FLO1 and other cell wall components underlie preferential aggregation with clonemates. In Aim 2, the researchers will assess how specific this recognition system is by testing whether independently evolved strains can distinguish between different genetic lineages. Confocal microscopy will be used to analyze the three-dimensional arrangement of cells within groups to determine if cells specifically recognize and bind to relatives at the cellular level. The results will provide mechanistic insights into how micro-scale interactions between cells underpin the origin of assortment at the level of multicellular groups, advancing our understanding of how genetically-diverse cellular aggregates make the transition to multicellularity. 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 · 2024-12

Deep neural networks (DNNs) have been a major driving force behind recent advances in data science and engineering. An emerging theme in DNN research is to exploit the intrinsic structure of the learning problems, such as symmetry, to improve the data-efficiency of DNNs in the small-data regime. Recent work on symmetry-preserving machine learning typically studies it in the ideal setting where the symmetry transformations are perfect, whereas in reality, however, they are usually “contaminated” by various sources of signal deformation. The aim of this project is to rigorously measure and guarantee the deformation robustness of general symmetry-preserving DNNs, as well as quantifying their resulting performance gain. Results of the research are expected to advance understanding of robust geometric deep learning, with a diverse range of applications from computer vision to scientific computing with limited data. The project will provide interdisciplinary training in applied mathematics, engineering, and data science to undergraduate and graduate students. The overarching theme of the project is to leverage mathematical tools from differential geometry, applied harmonic analysis, and applied probability to improve the statistical-efficiency of machine learning models. Special emphasis has been placed on the rigorous analysis and promotion of robust symmetry-preservation that is broadly applicable to arbitrary Lie group representations on general feature fields. In addition, the project aims to extend the idea of symmetry-preservation to deep distribution learning, and proposes a unified framework for data-efficient generation of distributions with intrinsic structures including—but not limited to—group symmetry; the improved statistical efficiency will be rigorously quantified through sample complexity analysis. The techniques to be developed in this project will be widely applicable across different disciplines, providing fundamental building blocks for the next generation of mathematical tools for the computational and geometric modeling of Big 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 2024 · 2024-12

This award will develop a multiscale experimental/modeling framework to obtain a fundamental understanding of the coupled ice/salt crystallization phenomena in low/zero clinker systems. Freeze-thaw causes billions of dollars of damage to concrete infrastructure and buildings in the US yearly. Impacts of climate change such as increased number and severity of freeze-thaw events are predicted to aggravate this damage. The socioeconomic consequences of these reductions in serviceability include impediments to economic activity and exacerbated inequities in access to quality infrastructure in marginalized communities. By exploring new materials with reduced environmental impact, this project will lead to the development of more resilient concrete formulations, significantly extending the lifespan of infrastructure. Project outcomes will open significant pathways for broad implementation of low/zero clinker systems in building and infrastructure applications. The project supports national interests by promoting scientific progress, improving infrastructure resilience, and reducing carbon emissions. It also enhances educational opportunities and STEM diversity through exposure of K-12 students and teachers to novel technologies and sustainability concepts. Furthermore, development of concrete sustainability seminars and advanced academic courses will lead to a more knowledgeable STEM workforce. The technical goals of this research are to elucidate the mechanisms of entrained air void formation/stabilization, saturation, and coupled ice/salt crystallization damage in low/zero clinker cementitious materials. Using a combination of multiscale experimental methods, molecular dynamics simulations, and advanced characterization techniques, the project seeks to understand the physico-mechano-chemical interactions at play. The project will develop a comprehensive multiscale experimental/modeling framework to study these interactions, linking microscopic characteristics to macroscopic performance. These findings will inform the creation of highly durable concrete mixtures suitable for cold environments. Ultimately, the project aims to produce a performance model to predict the longevity of low/zero clinker materials under freeze-thaw conditions, providing a pathway for their broader implementation in sustainable building and infrastructure applications. Microstructures of low/zero clinker systems could be engineered from the bottom-up to mitigate damage due to crystallization stresses. This award will advance the specific state-of-the-art in low/zero clinker systems and more broadly brittle porous materials from several scientific and technological perspectives. This research will also advance the knowledge base in material science, porous media mechanics, computational science, and advanced analytical and imaging techniques. 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.