Georgia Tech Research Corporation

NSF Awards · FY 2026 · 2026-08

Animals survive by making flexible decisions as they move through complex environments. They look around, gather information, change strategies, and act on internal goals, but it remains difficult to determine how the brain produces these behaviors. Modern recording technologies can now measure brain activity and behavior in great detail, creating an opportunity to study behavior in settings that are closer to everyday life than highly simplified laboratory tasks. This project will develop computational tools that connect natural behavior to brain activity and help answer not only what an animal did, but why it behaved that way. The work will advance basic understanding of how the brain supports flexible behavior and may improve future studies of neurological and psychiatric disorders, brain-machine interfaces, and artificial intelligence systems that adapt to changing conditions. The project will also create new educational opportunities through research-based courses, undergraduate participation in neuroscience and artificial intelligence research, mentoring for K-12 teachers and students, and broader access to interdisciplinary training. The research is based on the idea that natural behavior contains measurable evidence about the internal computations that support perception, decision-making, and action. It will pursue three independent and complementary threads. The first objective will discover sensory strategies from freely moving behavior by developing inverse reinforcement learning methods that infer goals and strategies from behavior and visual input. These methods will determine how visual scenes and recent behavioral history shape active sensing, reward seeking, and disengagement, and whether these strategies help explain activity in primary visual cortex. The second objective will focus on cognitive state modeling by developing neural-behavioral dynamical models. These models will test whether similar outward behaviors can arise from different internal cognitive processes when prefrontal cortex activity is analyzed together with behavior. The third objective will focus on motor and reward representation by developing continuous, compositional motor-policy models together with distributional inverse reinforcement learning. These methods will study how motor cortex and striatum encode reusable movement patterns, task-specific combinations of those patterns, and variability in rewards. Together, these threads will test whether behavior data can reveal mechanistic structure across perception, cognition, and action, yielding circuit-level biological insight that descriptive behavior models cannot provide. The project will evaluate the methods on existing collaborator datasets and public neuroscience resources, release software and tutorials, and integrate the research with undergraduate research experiences, K-12 outreach, teacher mentoring, new courses, and interdisciplinary workshops. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2026 · 2026-08

This Faculty Early Career Development Program (CAREER) award will be used to study the mechanics of interlocking granular materials that consist of hard, interlocking elements assembled into soft, deformable materials, and their ability to enable shape-shifting structures. Soft materials have been essential to innovations in tissue engineering, soft robotics, stretchable electronics, among many other applications; however, their mechanical behavior is hard to tune and scaling them for nano- to meter-scale applications (e.g., from miniaturized medical devices to meter-scale reconfigurable structures) is challenging. Since the underlying structure of interlocking granular materials can be precisely tailored using optimization and advanced manufacturing, they have higher potential to be engineered for targeted behavior than traditional soft materials. This research will address the key challenge of understanding the fundamental deformation and failure mechanisms of interlocking granular materials and develop computational models for design and analysis purposes. Furthermore, a partnership between the research team and a workforce development program at Georgia Tech will enable long-term immersion of high-school students into the research and provide a platform to pilot educational materials related to interlocking granular materials for release to the broader engineering education community. Interlocking granular materials lie at the intersection of traditional architected materials (e.g., truss lattices) with rigidly connected constituents and traditional granular materials (e.g., gravels) with disconnected constituents. Their mechanics differs from the former in that they can “flow” via relative movement of their particles and from the latter in that their ability to “flow” is limited by collision of the particles in tension as well as in compression. Topological interlocking is the key distinguishing feature that prompts a different theoretical and numerical treatment than in these other well-studied systems. Experimentation and expensive discrete element simulations have dominated the limited studies on interlocking granular materials, but a well-founded theoretical and computational mechanics description that facilitates integration of interlocking granular materials into methodical design strategies is not yet established. Inspired by continuum mechanics of soft materials, a two-potential approach will capture the effects of particle-level contact, deformation, and failure as a function of particle and network topology parameters, in a homogenized view of the discrete system, which will enable strategic patterning of interlocking granular materials into optimized structures with shape-shifting capabilities. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2026 · 2026-06

Modern cloud services rely on expensive and power-hungry hardware, making efficient use of computing resources essential for controlling cost and energy consumption. This project focuses on maximizing how much useful work each server can perform without becoming overloaded or unresponsive. The central idea is to make cloud systems determine, within a few microseconds, how much work a server can safely accept, allocate resources to individual tasks accordingly, and then distribute incoming requests across servers based on these allocations. Today, resource allocation and load distribution are handled independently, which leads to inefficient resource use and slow reactions to rapid changes in workload. By combining these operations into a coordinated framework, the project makes these capabilities easier for users to adopt. The overall goal is to improve cloud services without continuously adding more hardware. The project aims to redesign load and resource management in a coordinated manner across software and hardware layers. This problem is fundamentally challenging because resource demands vary widely across requests, bottlenecks shift over time, and independent control mechanisms often operate at similar timescales and interfere with each other. Addressing these challenges requires fine grained visibility into application behavior and new control abstractions that coordinate decisions across layers without introducing excessive overhead. To achieve this, the work is organized around three technical thrusts. The first thrust plans to develop unified and transparent mechanisms that track resource usage for each application request and enforce admission decisions across multiple shared bottlenecks. The second thrust plans to integrate these decisions with operating system scheduling, jointly managing application load and the resources allocated to handle it. The third thrust plans to extend these ideas to clusters of servers, redesigning load balancing, backpressure, and scaling mechanisms for applications built as chains of microservices. The broader impacts of this project include improved performance, lower cost, and reduced energy use for cloud services that support our modern economy. By reducing the need for over provisioned computing resources, the project contributes to more sustainable and environmentally responsible infrastructure. The resulting tools will help application developers build faster and more predictable systems without deep expertise in low level resource management. Educational activities will integrate the research into courses, seminars, and online programs, preparing students for careers in computing systems while encouraging innovation through hands on projects and research experiences. The project will release software, data, and experimental results through publicly accessible repositories and the accompanying website: https://saeed.github.io/career/ This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2026 · 2026-06

Solitary waves that maintain their shape as they propagate have been observed for nearly two centuries. Today, these coherent structures are known as solitons. They arise in many nonlinear wave models in mathematics and physics, and they play an important role in understanding how complex wave patterns evolve over long times. While many of their properties are understood in special exactly solvable models, their stability, interactions, and long-time behavior remain poorly understood in many physically important settings. Basic questions remain about when such structures persist, how they interact with radiation and with one another, when collisions are elastic or inelastic, and how complicated waves simplify over time. This project studies these questions through nonlinear dispersive equations, which provide a mathematical framework for wave propagation in a wide range of systems. By advancing the mathematical understanding of solitons and related coherent structures, the project promotes progress in the mathematical sciences and strengthens foundations that are broadly relevant to wave phenomena in applied science. The project provides research training opportunities for graduate students, postdoctoral researchers, and undergraduate students through mentoring, new courses, and a summer school. This project develops methods from partial differential equations, harmonic analysis, spectral theory, and dynamical systems to investigate the qualitative dynamics of nonlinear dispersive waves. It focuses on two closely connected directions. The first direction studies multi-soliton dynamics, including stability, global dynamics near multi-soliton configurations, and elastic and inelastic collisions. The second studies the asymptotic stability of moving kinks and related topological solitons in the presence of long-range scattering effects and internal modes. A longer-term goal is to combine these directions to analyze multi-kinks and multi-solitons under the joint influence of long-range interactions and internal resonances, and to move toward higher-dimensional topological solitons arising in mathematical physics. The investigator combines Strichartz estimates, distorted Fourier analysis, smoothing estimates, normal form methods, modulation analysis, and space-time resonance methods to build new analytical frameworks for stability, scattering, and long-time dynamics. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2026 · 2026-06

This research will focus on advancing the understanding of particle acidity in the atmosphere of cold environments and assessing the effects of fuel oil sulfur reductions on particulate sulfur and other species in Alaskan cities. A recent transition to lower-sulfur fuel oil as part of a government-mandated air quality mitigation strategy in Fairbanks and neighboring communities offers a rare opportunity to directly observe and quantify the atmospheric impacts of reduced sulfur emissions. This work advances public health and welfare by guiding national strategies to reduce exposure to fine particulate matter and addressing the unique air quality challenges faced by Arctic communities. The recently NSF-funded 2022 Alaskan Layered Pollution And Chemical Analysis (ALPACA) study in Fairbanks Alaska showed that a substantial portion of PM2.5 sulfur was found in sulfur compounds that are only formed via aqueous-phase chemistry within a narrow pH range. The unique Arctic winter conditions enabled a new understanding of aerosol processes that enhanced heterogeneous chemistry by partitioning gas-phase precursors (e.g., formaldehyde and sulfur dioxide) into the particle phase and influenced particle pH through the temperature-sensitive behavior of key semi-volatile species like ammonia. In this project, real-time monitoring instrumentation will be used to quantify PM2.5 sulfate, hydroxymethanesulfonate, and related sulfur species, while simultaneously characterizing aerosol thermodynamics focusing on liquid water content and acidity under Arctic winter conditions. The inclusion of high time resolution measurements of gas-phase ammonia, nitric acid, and other trace gases will greatly extend beyond ALPACA’s original scope to better understand the processes influencing aerosol composition and acidity. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2026 · 2026-06

This grant will provide partial travel support to about 15 U.S.-based students to attend the Doctoral Colloquium at the IEEE Visualization & Visual Analytics Conference (IEEE VIS) 2026. This is a single-day forum that fosters the academic and professional development of young researchers in the field of visualization. The VIS Doctoral Colloquium, founded in 2006, focuses on the participants’ doctoral dissertation research and aims to develop the intellectual rigor of their research program. The colloquium provides a forum for students to exchange research ideas, discuss their on-going work, and improve upon their research agendas. The research presented at the event reflects state-of-the-art in visualization. The Doctoral Colloquium will improve STEM education at the doctoral student level by providing students with targeted mentorship from experts in their professional community, at a critical stage in the students’ professional development. Each applicant will be reviewed by two out of the three Doctoral Colloquium co-chairs, assigned such that conflicts of interest are avoided. The review committee members make the decisions based on the application materials described in the VIS 2026 Doctoral Colloquium Call for Participation, based on evidence that (a) their previous research has made intellectual contributions to the field of visualization, and (b) their proposed research plans are at a stage in which they would benefit from feedback from the Doctoral Colloquium panel of distinguished researchers. Six colloquium panelists will be selected by the Doctoral Colloquium Chairs by July 2026. Panelists are chosen based on their reputation in the research field and their ability to provide constructive feedback to young researchers. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2026 · 2026-06

Machine learning is rapidly changing science, engineering, and technology, but many of its most successful methods require enormous amounts of high-quality data. This limits their usefulness in scientific settings where data may be expensive, noisy, incomplete, or difficult to obtain, such as turbulence modeling, molecular design, medical imaging, and the study of complex physical systems. This project will develop new mathematical foundations and computational tools that allow machine learning systems to make better use of limited data by incorporating structure, such as symmetry, low-dimensional patterns, and physical constraints. It will also develop methods for discovering hidden structure directly from data, including conservation laws and governing principles of dynamical systems. These advances will support more reliable, interpretable, and efficient artificial intelligence for scientific discovery, aligning with national priorities in artificial intelligence and strengthening the mathematical foundations needed for future scientific and engineering applications. The project will also contribute to education and workforce development through new courses, undergraduate research opportunities, data science bootcamps, open-source software, and outreach activities for students in grades 8 through 12, thereby broadening participation and preparing students for careers at the interface of mathematics, computing, and science. The investigator will study the two-way relationship between data and structure in machine learning. One direction will develop structure-informed learning models that incorporate known structural information, with emphasis on generative models and diffusion models that preserve symmetries, multimodality, low-dimensional structure, and other physically meaningful priors. The project will establish rigorous sample-complexity and computational-complexity theory to quantify when such structure improves learning, when the benefit saturates, and how structural priors interact with optimization and regularization. A second direction will develop data-driven methods for discovering unknown structure in dynamical systems, including conservation laws, integrability, and Lax pairs. These methods will combine neural-network-based learning, deflation strategies, symbolic regression, and theoretical analysis to produce interpretable models and principled guarantees. The expected outcomes include new mathematical theory, scalable algorithms, benchmark problems, open-source software, and practical guidance for designing reliable structure-informed artificial intelligence systems for scientific computing and 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 2026 · 2026-06

This proposal seeks to advance the fundamental understanding of Earth’s nitrogen cycle, with a focus on the land-atmosphere exchange of nitrogen. This research integrates field observations, land-surface and atmospheric modeling, and sensor development, establishing a foundation for large-scale monitoring and policy evaluation. The project will evaluate the use of low-cost reactive nitrogen measurements and machine learning methods for robust reactive nitrogen flux spatial interpolation. Comprehensive datasets and improved models will provide a robust scientific basis for evidence-based policy decisions in air quality regulation, agricultural nutrient management, and ecosystem conservation in the United States. This project synergistically combines direct flux measurements at "supersites" with inferential modeling that uses low-cost measurements and existing air quality observations. The objectives are to: (1) advance the fundamental understanding of atmospheric reactive nitrogen (Nr), including both reactive nitrogen oxides (NOy, sum of all oxidation products of nitrogen oxides (NOx)), total ammonium (NHx, sum of gaseous ammonia (NH3), and particulate ammonium (NH4+); (2) develop and implement innovative monitoring and modeling strategies for nitrogen management; and (3) cultivate the next generation of scientists and engineers while fostering public awareness of the importance of nitrogen management. The project includes a citizen science component that empowers students, teachers, agricultural stakeholders, and community volunteers to participate in scientific 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 2026 · 2026-06

Antibodies are key proteins of the immune system that help fight disease. In people, immune cells called B cells create antibodies and then evolves them. B cells take months to do this, which makes it difficult to study antibody creation and evolution. This CAREER project will design a method to create and evolve antibodies “from scratch” in yeast, which will open new avenues for exploring antibody creation, evolution, and function. The results of the project will provide insights into the formulation of more effective disease-targeting therapeutics. The project will support project-based course development and the creation of educational Apps to train students in synthetic biology and biotechnology. Inside a person, antibodies are created and evolved at massive scale (millions of new antibodies per day). It is the combined functionality of this ‘evolved antibody repertoire’ that is important towards fighting off disease-causing agents. Unfortunately, it is difficult to study antibody generation and evolution in a way that mimics how they occur naturally. Yeasts are excellent alternate hosts for antibody production because they can display antibody proteins on their surface. This project will fill this gap by engineering yeast to efficiently recapitulate key aspects of recombination and antibody affinity maturation. The project will employ cellular engineering, protein targeting, synthetic biology, and antibody yeast display. Multiplex genetic engineering and enzyme localization studies will be employed to identify the key B cell and native yeast cell enzymes that enhance yeast-mediated recombination, which is a DNA cleavage and repair process. Bespoke selection strategies and genetic arrays will allow for generation of, and controlled selection for, functional antibodies made from scratch in yeast cells. Further incorporation of highly efficient and targeted in vivo mutation of antibody genes will allow for analyses of the evolution of antibody function. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2026 · 2026-05

This project is funded through the NSF Translation to Practice (TTP) program, which supports efforts to turn research discoveries into practical tools that benefit communities, industry, and society. Solar storms can knock out the radio communications that aircraft, ships, and emergency responders depend on — sometimes for hours at a time, with serious safety and economic consequences. A key reason these outages are difficult to predict and manage is that a particular layer of the atmosphere that lies at the edge of space, called the D-region, remains largely unmeasured and poorly understood. Sitting roughly 40 to 55 miles above Earth's surface, the D-region is too high for weather balloons and too low for satellites to reach directly. This TTP-P award supports a team at Georgia Tech, in partnership with the weather technology company Vaisala Inc., to build a system for mapping the D-region continuously and in near real time across the entire globe. The team uses radio signals produced by lightning — which strike millions of times each day worldwide — to detect and track conditions in this difficult-to-reach atmospheric layer, in a way that is similar to how doctors use magnetic resonance imaging (MRI) technology to see inside the body. Better knowledge of the D-region will help airline and maritime operators manage radio communication during solar storms, help power grid operators prepare for damaging solar disturbances that can cause blackouts, and strengthen backup navigation and timing systems in case a Global Positioning System (GPS) becomes unreliable. The ionospheric D-region (60–90 kilometer altitude) is the primary absorber of the high-frequency radio signals used by aviation, maritime, military, and emergency communications. It is also the atmospheric layer most directly affected by solar X-ray emissions during space weather events. Despite its importance, real-time D-region monitoring remains limited. A leading operational product — National Oceanic and Atmospheric Administration's D-Region Absorption Prediction (D-RAP) — relies on empirical models with coarse spatial resolution and no direct measurement of electron density. This project advances toward commercialization a D-region tomographic imaging system developed at Georgia Tech, which uses very-low-frequency (VLF) radio signals from both natural lightning and ground-based radio transmitters to produce spatially resolved, three-dimensional maps of D-region electron density. The core innovation is a physics-constrained inversion algorithm — drawing on techniques from medical imaging — that fuses these two complementary observation types into global maps that can be updated on minutes time scale, a horizontal resolution of a few hundred kilometers. The research and translational activities are carried out in partnership with Vaisala Inc., whose global lightning detection network already includes dozens of operational VLF receivers distributed worldwide, enabling global coverage without new hardware deployment. Research objectives include reformulating the mapping algorithm for a spherical Earth geometry, achieving near-real-time operational performance in Vaisala's computing environment, and validating results against other independent measurements. Anticipated outcomes include a validated global D-region mapping prototype with quantified uncertainty at every grid point, a commercialization plan centered on technology licensing to Vaisala, and open algorithmic tools for the scientific community. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2026 · 2026-05

The 2026 Hilton Head Solid-State Sensors, Actuators and Microsystems Workshop (HH2026) is a prominent conference in the field of microsystems, particularly the micro-electro-mechanical systems (MEMS) microsystems, as well as their applications. It is the 22nd in the series of Hilton Head Workshops and is expected to draw close to 500 attendees with various engineering and scientific backgrounds (electrical engineering, mechanical engineering, materials science and engineering, biomedical engineering, chemical engineering, chemistry, physics, biology, etc.) from academic, industry, and government. HH2026 will have a focus on design and manufacturing of new devices and microsystems including sensors, actuators as well as emerging applications in biosensing, computing, and communication. Special topic sessions will further focus on AI and machine learning for digital twins and MEMS engineering. These cutting-edge research areas are aligned with NSF priorities in AI, microelectronics and semiconductor, advanced manufacturing, and biotechnology. The Hilton Head Workshop will be held in Hilton Head Island, South Carolina on May 31 - June 2, 2026. To support the training of the next-generation scientists and engineers working in this critical research field, this NSF travel grant will provide travel support for 25 U.S. students or post-doctoral researchers to attend and present their papers at the workshop. Their participation in the workshop will enable them to learn the newest research and development trends in MEMS microsystems as well as provide them excellent networking opportunities through various events to interact with senior researchers in this field. 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 2026 · 2026-05

This CAREER will advance the understanding of flows involving liquid and gas mixtures that form bubbles, droplets, and sprays in nature and technology. These flows appear in ocean waves, clouds, rainfall, and lungs, as well as in chemical reactors, pipelines, and energy systems. Predicting how bubbles and droplets form, merge, and break apart is these flows is challenging. These phenomena are influenced by fast and complex interactions that occur across a wide range of sizes and speeds. They are not well represented in current computer models. This project will develop improved tools for simulating these flows by focusing on how the surfaces between gas and liquid stretch, wrinkle, and change over time. Better simulation tools will benefit scientific discovery and support advances in national priority areas such as energy production, chemical processing, weather forecasting, transportation, aerospace, and naval systems. The project will also help build a skilled engineering workforce through integrated education and outreach efforts. High school, undergraduate, and graduate students will participate in hands-on research and computational training, strengthening pathways into science and engineering careers. This award will develop a predictive simulation framework for gas-liquid flows characterized by complex interfacial dynamics. The research will investigate how interfacial surface area is generated and destroyed during surface wrinkling, breakup, and coalescence, and how these topological changes influence the exchange of mass, momentum, and energy across the interface. The project will introduce a modeling framework that treats interfacial area as a dynamic transported quantity and develops multiscale models grounded in first-principles analysis and high-resolution numerical simulations. The work will combine three components: (1) quantitative assessment of multiscale interfacial geometry; (2) development of sub-resolution models that capture interfacial topological changes; and (3) application of the framework to non-equilibrium flows relevant to energy and environmental systems, including bubble columns, turbulent duct flows, and liquid sprays. Validation will be performed using experimental measurements and high-fidelity simulations, enabling model improvement and assessment of predictive capability. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2026 · 2026-05

This project will train the next generation of seagoing physical oceanographers through a field campaign in the South Atlantic Bight and adjacent Gulf Stream system. Under faculty mentorship, students will design and execute a sampling strategy to investigate shelf–Gulf Stream exchange and upper-ocean variability along the continental margin. The program integrates pre-cruise planning, shipboard data collection, and post-cruise data synthesis, providing participants with experience in observational oceanography and field campaign design. It expands access to ship-based training in physical oceanography to all types of institutions. These experiences equip participants with the technical and operational skills needed to contribute to ocean observing systems, field campaigns, and data-driven research. Cruise data and instructional materials will be made publicly available, enabling broader use in teaching, research, and regional ocean observing efforts. The cruise will focus on observing physical processes associated with continental shelf and Gulf Stream interactions. Students will design and implement a targeted sampling plan to capture features such as frontal structure, stratification, and shelf-break exchange. Participants will gain practical experience deploying core physical oceanographic instrumentation while learning to adapt sampling strategies under operational constraints. Cruise observations will be processed, quality-controlled, and analyzed withing standard oceanographic dynamical frameworks. Data collected during the cruise will be archived in the Southeast Coastal Ocean Observing Regional Association repository, contributing to regional observing capacity. Students will also be introduced to the basics of regional ocean modeling, including how to configure and run a nested model domain representative of the observed region. Real-time modeling exercises informed by observations collected during the cruise will allow students to directly connect live field measurements with numerical model outputs. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2026 · 2026-05

The National Nanotechnology Coordinated Infrastructure (NNCI) comprises 16 university sites and their partners, funded by the National Science Foundation starting from 2015-2024, to provide open access to nanoscale fabrication and characterization user facilities and staff expertise across the nation. The Coordination of Nanotechnology Research Facilities is designed to facilitate and provide continuity of the NNCI as it transitions to a new network, the National Quantum and Nanotechnology Infrastructure (NQNI). The goal of this proposed effort is to maintain a national network that catalyzes discovery in nanoscience and nanoengineering and helps transform these discoveries into technologies and products that address global grand challenges. This objective is enabled through programs of user and site support, education and outreach, societal and ethical implications, computation, innovation and entrepreneurship, and helps provide linkages with other national and international nanotechnology resources. The activities of the NNCI contribute to the economic competitiveness of the U.S. in training a globally competitive workforce and in providing efficient access to resources for commercialization of nanotechnology. They also help to inform and educate the general public on fundamentals and advances of nanoscience and engineering and their societal impact. The objective of the Coordination of Nanotechnology Research Facilities is to strengthen and accelerate discovery in nanoscale science and engineering throughout the US and world by supporting the continuing goals of the National Nanotechnology Coordinated Infrastructure (NNCI) to provide access to enabling infrastructure and staff expertise. The network working groups and research communities will be supported to solve common problems, share best practices, and address specific research areas. Workshops, symposia, and staff exchange activities will be funded to encourage collaboration. Development of materials, tools, and visualizations promoting and enabling wider use of simulation and education tools for the design, fabrication, and characterization in the area of nano/quantum technologies will be integrated into a tool-based curriculum hosted on the nanoHUB platform. The network’s impact beyond the US will be enhanced through participation in Global Nanolab, an international consortium of nanofabrication networks recently initiated and designed to leverage collective expertise, share best practices, and support technical staff interactions. The NNCI network will continue to contribute to the economic competitiveness of the US by training a globally competitive workforce as well as innovators and entrepreneurs skilled in translating nano-enabled technologies for societal impact. The undergraduate student, graduate student, and industry users of the research facilities achieve skills in the techniques of nanoscale science and engineering resulting in professionals from all backgrounds who are ready to meet the global workforce demands of the 21st century. The education and public outreach activities coordinated through the network encourage K-12 students to participate in the STEM pipeline and help create an informed citizenry that supports the safe development of nanotechnology. An annual REU Convocation will help summer research interns develop communication and presentation skills and network with their peers. In addition, societal and ethical implications workshops are designed to enable nanoscale scientists and engineers to interact with social scientists and policy experts so that the potential impacts new discoveries might have on society are considered. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2026 · 2026-05

Visual impairment is a major health issue that affects the quality of life for many people. This project will develop a sensory augmentation system that will help people navigate complex environments. The system of devices will use the physical sense of touch to enable people to “feel” shapes and structures of surrounding objects. The project will integrate the devices into breathable, everyday clothing to deliver touch cues. These cues will indicate obstacles and guide users toward objects of interest. The clothing will be designed for adaptability, cost-effectiveness, and user-friendliness. The project will advance fundamental knowledge in wearable electronics, perception, and artificial intelligence. This project will also build talent in engineering by developing practical, lab-based experiences for undergraduate students and by introducing new graduate coursework for graduate students. Skin presents unique challenges for delivering clear, reliable touch-feedback across large areas. To address this, the team will develop a wireless mesh network of modular haptic units and integrate them into everyday clothing. The system will deliver rich touch feedback that approximates natural tactile sensations. Bistable, bioelastic operation will keep the devices compact and energy efficient so this feedback can be distributed across the body. Objects in the user’s surroundings will be detected using cameras and time-of-flight sensors. A large-language model will translate plain-language instructions from the user for on-demand personalization. This code will define how haptic feedback responds to obstacles and points of interest detected by the connected sensors. Principles of active perception will align feedback with natural movements, making it more intuitive and reducing training requirements. Feasibility studies with blind and low vision people will evaluate functional performance, user preference, comfort, and affective responses. This project will examine how users can “feel” obstacles and waypoints using the haptic system as they navigate and engage with their surroundings. Overall, this project introduces a new AI-enhanced, user-driven strategy for touch-based navigation. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2026 · 2026-04

High-Level Synthesis (HLS) is a design methodology that allows developers to create hardware systems, such as accelerators for artificial intelligence and data processing, using common programming languages. HLS improves productivity and accessibility, but fragmented tools, inconsistent evaluation practices, and the absence of shared, reproducible datasets hinder progress in HLS research and education. This project addresses these challenges by creating an open-source ecosystem that unifies tools, workflows, and datasets for HLS-based hardware design. By lowering technical and expertise barriers, the project enables students, software developers, and early-career researchers to participate helping to increase access to hardware innovation. The ecosystem promotes open science, reproducibility, and fair comparisons across platforms, strengthening national research infrastructure in computing and semiconductor design. Integration into university courses, capstone projects, summer schools, and online training programs support workforce development in advanced computing fields that are critical to national competitiveness. Although High-Level Synthesis (HLS) is increasingly used for hardware design, research in this area remains difficult to reproduce and compare due to fragmented toolchains, incompatible benchmarks, and inconsistent evaluation metrics. In addition, challenges are further amplified for machine learning and large language model (LLM)-based approaches. This project addresses these gaps by expanding an existing open-source framework into a modular open-source ecosystem for reproducible benchmarking, systematic design space exploration, and data-driven HLS research using machine learning (ML) and LLMs. The ecosystem provides unified interfaces for coordinating multiple HLS toolchains, standardized quality-of-result metrics across different Field-Programmable Gate Arrays (FPGAs) and Application-Specific Integrated Circuit (ASIC) targets, and scalable pipelines that generate large, structured datasets from limited seed designs. The project also supports consistent evaluation of automated and LLM-generated designs and offers extensible application programming interfaces for integrating new tools, hardware backends, datasets, and learning models. This ecosystem establishes a rigorous and adaptable foundation for reproducible, ML-enhanced hardware design using HLS. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2026 · 2026-03

This project provides funding for students from U.S. institutions of higher learning to attend the 2026 IEEE International Conference on Pervasive Computing and Communications (PerCom). Pervasive computing focuses on technologies that are critical for many areas of high importance to society, such as smart cities, connected communities, efficient transportation networks and secure electrical grid. These technologies enable the development of novel AI-based applications using inexpensive resource-constrained software and unreliable networks, allowing wide deployment across cities and communities. Increased participation of US-based undergraduate and graduate students in PerCom would significantly contribute to the development of forward-looking CPS workforce. It provides students with the opportunity to interact with more senior researchers, and exposes students to emerging ideas and current industry practices in this important area. PerCom 2026 is organized in March 2026 in Pisa, Italy. In addition to the main conference, it includes a large number of workshops and networking-oriented activity, including the Ph.D. Forum for graduate students. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2026 · 2026-02

Non-Technical Abstract: Harnessing quantum materials for cutting-edge scientific research and technological innovations constitutes a key mission of the ongoing quantum revolution. Magnetic Weyl semimetals, a group of scientifically intriguing and technologically important materials, are naturally relevant in this context thanks to their promise for delivering advanced functionalities such as high density, ultrafast processing speeds and improved stability to next-generation quantum electronics applications. In this project, the principal investigator plans to introduce scanning-probe quantum microscopy to investigate the underlying correlation between microscopic magnetic phases and topological properties of emergent Weyl semimetals. An important goal here is to leverage quantum spin sensors to reveal the interplay between topology, magnetism, and valley transport behaviors in topological quantum magnets. In parallel with the proposed research, this project also dedicates a major effort to increasing society’s awareness of some of the most exciting developments and challenges in condensed matter physics, materials science and quantum sciences and technologies. It promotes the participation of young researchers at the forefront of interdisciplinary scientific research. Outreach activities include lectures, learning and demo materials for nearby technical colleges and high schools, so that contemporary scientific knowledge can reach out to a broad audience. Technical Abstract: Recently, magnetic Weyl semimetals have received immense interest in cutting-edge condensed matter physics research. Tremendous efforts have been devoted to investigating exotic topological behaviors in this new family of quantum materials. Examples include Weyl fermion mediated ferromagnetism, chiral magnetic effect, magnetic vortices, quantum anomalous Hall effect and many others. Here the principal investigator proposes utilizing quantum spin sensors to perform nanoscale quantum sensing of novel magnetic and noncentrosymmetric Weyl semimetals RAlSi (R = rare earth element). Exploiting the highly competitive spatial and field sensitivity of scanning-probe quantum microscopy, this project aims to visualize the exotic spin density waves and to probe valley transport properties in NdAlSi, uncovering the underlying mechanism of Weyl mediated unconventional magnetism. Taking advantage of the dipole-dipole coupling between quantum spin sensors and Weyl materials, the research team further proposes to achieve local excitation, control, and detection of axial electromagnetic field-driven chiral anomaly in CeAlSi, opening new pathways to explore microscopic quantum spin and charge behaviors in topological materials. This project makes important contributions to the burgeoning field of quantum materials and promotes the role of Weyl semimetals in developing innovative quantum microelectronic devices for next-generation information technologies. By developing scanning-probe quantum sensing techniques and demonstrating their operation over a broad range of conditions, it also brings new experimental approaches to study topological physics in emergent quantum states of matter. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2026 · 2026-01

This National Science Foundation Innovations in Graduate Education (IGE) Track 2 award to the Georgia Tech Research Corporation will examine how combining industry support and internships with traditional support mechanisms, such as assistantships and fellowships, can improve the educational experience for STEM Ph.D. students. This project also explores how models of industry engagement in Ph.D. students' educational pathways influence faculty members, including how faculty members shape their research priorities, mentorship practices, and grant-seeking behavior. By studying how industry-engaged models influence mentorship, collaboration, and long-term success, the project aims to uncover ways to improve graduate education for students knowing that the students' experiences should align with potential careers in both academia and industry. The findings will provide a blueprint for implementing scalable practices that reshape how universities support graduate students and foster stronger connections between academia and the workforce. The project will use a mixed-methods research design to evaluate the effectiveness of industry-engaged flexible educational engagement models in STEM Ph.D. programs. It will focus on two key areas: mechanisms of support and graduate research environments. Researchers will collect quantitative data through surveys of students, faculty, and industry partners, examining variables such as career outcomes and research productivity. In-depth qualitative insights will be gathered through semi-structured interviews with stakeholders, including students, faculty mentors, industry collaborators, and university administrators. Longitudinal data from institutional records will track student progress and career trajectories over time. The study will seek to understand how industry engagement and other structures influence motivation, engagement, and success. In its first year, the project will focus on developing research instruments and the recruiting of participants. Data collection will span years two through four, with analysis and refinement occurring throughout. In the final year, the team will disseminate findings, release institutional policy recommendations, and evaluate the broader impacts of the work. Expected outcomes include evidence-based strategies for implementing sustainable, scalable industry-engagement models, improved understanding of how industry partnerships shape graduate education, and actionable insights for industry and career opportunities 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 2026 · 2026-01

The use of artificial intelligence (AI) is becoming of paramount importance for scientific discovery. Scientific AI projects rely upon complex collaborative workflows combining diverse (and potentially sensitive) data sources and artifacts, and as such require reproducibility, accountability, and transparency. Alterations of the AI data or models (either at collection time, at training time or at inference time) pose a severe threat to the accuracy and therefore usability of outcomes from scientific AI. In theory, some of these concerns can be mitigated via federated learning which supports collaborative learning, and cryptographic techniques. However, there has been neither systematic exploration nor toolchains of techniques that can be leveraged and deployed in real systems to serve real-world AI-enabled scientific applications. This project combines techniques from AI, systems security, cryptography and privacy to develop secure and reliable methodologies---targeted to the use of AI in medicine---for data and model provenance tracking in scientific AI applications. In more detail, the project works along two axes underlying integrity, provenance, and authenticity (IPA) of AI scientific applications, namely data (training and querying) and AI models. The starting point is a systematic extension of the associated system infrastructure to support more functional and certified logs—in order to enable reproducibility. Such extensions enable embedding of highly optimized cryptographic techniques to ensure the IPA of data and models. The novel system extensions are leveraged to design and implement new cryptographic techniques that ensure IPA of data---e.g., hash-chains for reordering prevention and watermarking for statistically verifiable provenance---as well as AI models---e.g., zero-knowledge proofs for certifying model ownership and privacy estimators for detecting dependencies. Using medical research as the project’s motivation and main application and tailoring solutions and toolchains to it offers new avenues to collaborative scientific discovery in this highly sensitive application domain. More generally, the project’s novel systems extensions create a platform for cryptography research to integrate further advanced solutions into the AI ecosystem. As such, the developed tools enable cross-disciplinary collaborations between areas whose technical depth makes it increasingly challenging to train experts on their intersection—namely AI/ML and cryptography. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

Satellites serve as the backbone of our digital world, enabling global communications, navigation systems, weather forecasting, and internet connectivity in remote areas. They facilitate everything from GPS navigation in smartphones to international financial transactions and military operations. When satellite systems fail, the consequences are immediate and far-reaching: navigation systems become unreliable, affecting transportation and logistics; communication networks experience outages, particularly in remote areas dependent on satellite internet; weather forecasting accuracy diminishes, impacting disaster preparedness; and many financial and commercial systems relying on precise timing signals face major disruptions. The ability to mitigate failures and enable routine repairs and hardware upgrades of spacecraft in orbit hinges on advanced space robotics capabilities. However, the harsh environmental conditions of space—with challenging illumination, limited computational resources, and diverse motion regimes—make traditional computer vision and localization techniques developed for terrestrial applications inadequate for space applications. Space robotics demands perception capabilities that far exceed those of similar terrestrial systems. This research develops novel visual perception, localization, mapping, and planning algorithms to enable new capabilities for all future space missions such as failure mitigation, large flexible structure assembly, orbital debris removal, inspection, hardware upgrades, and many more. This research involves training of both graduate and undergraduate students. The results of this research are disseminated to the community by journal and conference publications, organization of invited workshops and seminar presentations, and by targeted exposure (press releases, interviews) to popular media. This research advances the state-of-the-art in computer vision and perception for space applications by addressing three critical limitations: first, existing feature detection methods struggle with the harsh illumination conditions of space; second current 3D reconstruction techniques fail to account for the dynamic orbital environment; and third, traditional simultaneous localization and mapping (SLAM) approaches cannot handle the diverse motion regimes of resident space objects (RSOs). To address these challenges, the research team develops a robust learning-based feature detection framework for space by using line features that exploit the inherent geometry of the target, along with multi-spectral imaging feature extraction. It also develops a novel dynamics-aware 3D Gaussian Splatting framework that incorporates relative orbital dynamics as physical constraints, enabling simultaneous state estimation and motion regime classification while maintaining physical consistency. The research team introduces a visibility-aware neural field representation that explicitly models observation uncertainty to drive information-theoretic view planning, therefore enabling autonomous space robots to systematically explore unknown objects through efficient observation sequences. Finally, the developed theory is experimentally tested and validated using Georgia Tech’s Autonomous Spacecraft Testing of Robotic Operations in Space (ASTROS) platform, a state-of-the-art spacecraft simulation platform, and also using a high-fidelity synthetic simulation environment. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

This project aims to develop methods to create individually tailored large language models (LLMs) in the context of systems for strengths-based coaching of job seekers. People with autism have a wide range of abilities and needs around the interactions needed to conduct a job search. Strengths-based approaches to job coaching, which identify and apply each person's unique strengths in preferred work environments, show promise in helping individuals secure competitive employment, but the high cost and labor involved make strengths-based coaching inaccessible to many young adults. Currently, individuals turn to tools like ChatGPT to help with cover letters, interview preparation, and other job-seeking tasks. However, these tools often produce generic responses that fail to reflect the unique strengths of job seekers and raise ethical concerns about handling sensitive, diagnosis-related information and the risk of over-reliance on the LLMs. Through developing new ways to build personalized LLM-based agents that discover and incorporate people's strengths and needs, this project will both advance the utility and safety of LLMs and increase the employment projects for individuals with autism and others, benefiting society. To address these challenges, the project will follow a three-part approach. First, the project team will analyze chatbot dialogue data and neurodivergent user profiles from the LLM job coach feature currently deployed on a neurodiversity-focused employment platform that connects neurodivergent job seekers, including autistic individuals, with employers. Insights from this analysis, along with input from individuals and professional job coaches, will guide the development of job coaching guidelines, identify opportunities for strengths-based support, and uncover ethical risks and safeguards to inform the design of a strengths-based job coaching LLM model. Second, the project team will create a two-part system based on the identified design guidelines: a fine-tuned LLM for job coaching and a control model that uses individual strengths to generate personalized, context-specific responses. This design addresses the limitations of generic, one-size-fits-all outputs. Third, the project will build a novel job coaching system leveraging these advances and test how effectively the developed strengths-based job coaching LLM can support job seekers with autism and others. Through co-design workshops and lab-based evaluations, the project will assess the system's ability to improve people's confidence, self-esteem, and perceived usefulness of the coaching process, while identifying areas where the model may need refinement to ensure ethical and practical implementation at scale. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

The construction industry is not only one of the industries with the highest fatality rates, but also significantly consumes natural resources, with nearly $2.1 trillion worth of buildings constructed annually, resulting in a substantial ecological footprint. To address this, the construction industry is leveraging sensing technologies to achieve sustainable construction practices, making it necessary to upskill the future workforce in this area. However, the dynamic nature of construction sites and the ubiquitous safety challenges often impede hands-on experiences in practicing the use of sensing technologies to solve real-world problems and achieve sustainable construction practices. This project will investigate an embodied virtual reality and artificially intelligent environment (VRAPS) to facilitate the development of problem-solving skills of heterogeneous learners for sustainable construction. VRAPS will also be designed as an intelligent platform that can adapt to the diverse needs of different learners, making it a transferable learning environment with distinct features. This project will demonstrate how a personalized learning environment can close the competency gap between industry and construction education by developing a workforce capable of advancing sustainable practices in the construction industry. This planned research investigates a virtual reality and artificially intelligent environment for personalized learning in sustainable construction education (VRAPS). Through a mixed-methods approach, the project aims to answer the following questions: What competencies are required to advance sustainable construction using sensing technologies? What characteristics of an embodied virtual reality environment facilitate personalized learning of sustainable construction? And to what extent does a personalized and embodied virtual reality environment enhance problem-solving skills that are needed for advancing sustainable construction education? The first research question will be addressed through a Delphi study and a focus group discussion with industry practitioners to understand the competencies required to advance sustainable construction. The results will inform a formalized construction-domain learning content, which will drive the development of VRAPS. The proposed research will then combine interactive tools, technologies, and techniques, such as sensing technologies, virtual learning assistant, tangible objects, and artificial intelligence, to develop VRAPS. The development of VRAPS will provide engaging learning tools to complement traditional instruction, allowing teachers to tailor instruction to students’ needs. The second research question will be answered by leveraging behavioral data to investigate the characteristics of VRAPS that facilitate personalized learning. To answer the third research question, the research team will implement and assess VRAPS in two institutions to understand its effectiveness in enhancing the problem-solving skills required for advancing sustainable construction. VRAPS will contribute to the active learning theory through the use of multimodalities (such as virtual learning assistant and tangible interfaces) to support continuity during the acquisition of problem-solving skills for sustainable construction education. To reach a broader audience, VRAPS will be implemented for professional development, summer camps and outreach opportunities at Georgia Tech and CSU Fresno. This project is funded by the Research on Innovative Technologies for Enhanced Learning (RITEL) program that supports early-stage exploratory research in emerging technologies for teaching and learning. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

Recent advancements in Artificial Intelligence (AI) drive unprecedented innovation but face the critical challenge of escalating power consumption. The rapid expansion of AI leads to unsustainable energy use for computation and cooling, threatening its widespread adoption. While three-dimensional heterogeneous integration offers performance and energy efficiency gains through miniaturization for data-intensive AI workloads, this miniaturization simultaneously increases power density and reduces thermal conductivity. This creates severe localized hotspots that significantly degrade system performance, efficiency, and reliability, negating an estimated 40% of potential gains from each technology generation. The research team addresses this crucial barrier in AI and computing. The intellectual merits of this effort lie in an unprecedented, convergent research that integrates expertise in circuits and architectures, optimal control, and thermal transport and modeling. Advances in circuit design, thermal modeling, and control that are enabled by this research converge to realize, for the first time, a comprehensive framework for thermal management. The broader impact of this research extends beyond its potential to alleviate thermal challenges, and thus paving the way for continued technological advances in computing. By capturing the increasingly critical interactions between the computational and physical state of the system, the open-source Crucible simulation tool represents a critical framework for thermal management across a broad range of systems. Both the relevance and impact of this tool are magnified through active and continuous engagement with the semiconductor industry. This research also contributes to broader educational advancement of undergraduate and graduate courses, dissemination of results, and development of an open-source infrastructure for run-time cyber-physical system management for nationwide student use. Current design-time and run-time techniques for mitigating hotspots do not scale effectively to larger systems and offer limited capabilities for sensing and actuation to maintain thermal compliance. As a result, these methods are ill-suited for modern heterogeneous three-dimensional heterogeneous integration systems. There is currently no existing mechanism that can analyze or simulate these systems in a closed-loop manner, providing accurate digital-physical modeling that precisely reflects the run-time impact of controller actions on system function, performance, and thermal properties. To address this critical gap, the research team proposes run-time optimal thermal management for three-dimensional heterogeneous integration. This approach seeks to maximize a given performance objective while adhering to system-wide thermal constraints. Instead of more traditional millisecond-scale control of voltage and frequency, this project aims to achieve microsecond-scale sensing, actuation, and control circuitry. The research team will achieve this rapid response by combining a fine-grained network of thermal and load current sensors with an accurate reduced-order predictive thermal model and hierarchical model predictive control at runtime. This effort represents a novel confluence of techniques from thermal modeling, circuit design, and control. The proposed approach minimizes the need for conservative thermal margins, thus unlocking and releasing significant performance and efficiency gains with each new technology generation. A significant outcome of this research will be the creation of Crucible, a tool-agnostic simulation framework. Crucible relies on the existing capabilities of widely available switch-level simulators to jointly model both the digital and physical characteristics of the closed-loop control system at user-defined timescales. Crucible is anticipated to facilitate methodical exploration into run-time system optimization that requires integrated digital-physical modeling and control. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

Recent advancements in large machine learning models have demonstrated that increasing the number of parameters enhances computational precision and unlocks capabilities once deemed unattainable. This trend is exemplified by the rapid growth in model sizes, for instance, GPT-3 contained 175 billion parameters, while GPT-4 reportedly utilizes up to 1.8 trillion. This trajectory is expected to continue in the foreseeable future. However, the explosive growth in model size presents two major challenges for computer architecture and systems research: prolonged simulation times, which can extend from several days to weeks for large-scale models, and infeasibility of deploying workloads on a single compute engine (e.g., a graphics processing unit (GPU)) due to limited on-device memory capacity. To address these challenges, this project proposes the development of scalable simulation techniques and advanced memory management strategies tailored for large-scale machine learning workloads on GPUs. Unlike existing application-agnostic approaches, this research will leverage the distinctive data access patterns and value distributions of modern machine learning models to enable more efficient memory compression and more accurate simulation acceleration. While the primary focus will be on emerging machine learning models, the broader objective is to advance GPU computing to better accommodate any big data workload constrained by memory limitations. This will facilitate faster and broader adoption of GPUs across diverse computing domains, driving continued innovation in computational science. The outcomes of this research will be integrated into both new and existing undergraduate and graduate curricula, as well as K-12 outreach initiatives, fostering a deeper understanding of cutting-edge computing technologies across educational levels. This project would answer two research questions: how to simulate large machine learning computing and how to utilize GPU local memory better when the memory is oversubscribed. While large-scale simulation and memory management have been widely studied, most existing approaches fail to capture the unique architectural characteristics of GPU computing and the specific behaviors of emerging machine learning workloads. Rather than relying on application-agnostic or user-dependent sampling techniques, this research will exploit the distinctive compute and memory access patterns inherent to machine learning models. The first thrust will research efficient simulator acceleration methodology by leveraging the fact that machine learning models are typically executed with highly optimized library functions. These library functions tend to have similar architectural behaviors depending on the operational and data size characteristics. The project will identify representative sample kernels whose performance can be extrapolated to other similar kernels, thereby significantly reducing simulation overhead. By leveraging characteristics of the library functions, the second thrust will explore efficient memory expansion and compression strategies such as dynamic memory prefetching and eviction policies to mitigate the effects of memory oversubscription. The second thrust will develop novel quantization techniques that take advantage of the unique value distributions of weights and gradients within individual tensors. Unlike tensor-oblivious methods, this targeted approach aims to reduce memory footprint more effectively while preserving model accuracy. 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.