University Of Massachusetts Amherst

NSF Awards · FY 2026 · 2026-07

Artificial intelligence (AI), large-scale data, connected devices, cloud and high-performance computing systems, that together form the nation's cyberinfrastructure, are central to national competitiveness, yet most computing students encounter them only in advanced elective courses. This project infuses aspects of AI, Big Data, and Parallel and Distributed Computing concepts and practices into three foundational computing courses. The first two form the usual introductory programming sequence, and third is the computer systems course that is often taken shortly after them. Thus, all computing majors, not only those pursuing upper-level elective courses, will develop critical skills for understanding and contributing to the modern computational ecosystem. The project addresses a persistent barrier to such curriculum modernization: many instructors need focused preparation, classroom-tested examples, and adaptable teaching materials before they can confidently introduce these topics in early courses. To overcome this bottleneck, the project develops courses and materials to train about 200 current and future instructors through three intensive in-person summer workshops, an additional six hybrid tutorials at major conferences, and complementary online workshops. Summer trainees adapt and implement the course exemplars at their own institutions and contribute evaluation data, classroom-tested refinement and local adaptation, enabling broader adoption. With the potential to impact about 250,000 students over 5-10 years, the project serves NSF's mission by strengthening computing education, expanding access to AI and advanced cyberinfrastructure skills, and building the nation's long-term technological and research workforce capacity. The project advances knowledge in computing education by producing rigorously classroom-tested exemplars infused with Artificial Intelligence (AI), Big Data (BD), and Parallel & Distributed Computing (PDC) for Computer Science 1 (CS1), Computer Science 2 (CS2), and Computer Systems courses. Implementation across 60 diverse institutions will generate evidence-based models that can be widely adopted, thereby transforming early computing education at scale. The project investigates how AI-enabled learning tools and pedagogy can modernize core curricula by enabling students to construct, explore, and reason about modern computing systems earlier and in more depth than was previously possible. Evaluation data from trainees' implementations - including student learning, retention, and institutional adoptability - will contribute to generalizable knowledge on the design and scaling of Cyberinfrastructure-centric curriculum innovations. The project incorporates aspects of AI, BD, and PDC concepts and practices into the foundational computing courses ensuring that all computing majors, not only those pursuing upper-level electives, develop critical Cyberinfrastructure-ready skills. The project's three core course exemplars will be nationally adoptable, with trainees providing local adaptations to build a community-driven ecosystem of shared materials. Two key innovations in this project are: (i) harnessing AI both for pedagogy and for enabling course modernization: AI tools will make it possible for introductory students to develop, experiment with, and understand software, data, and other artifacts that were previously too complex to explore meaningfully at scale; and (ii) explicit integration of how BD and PDC power AI, giving computing students insight into the Cyberinfrastructure ecosystems underlying AI-driven discovery and innovation. 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 · 2026-07

Modern software increasingly relies on autonomous AI-driven systems that generate, test, and revise code. Yet even the most advanced tools struggle with a simple question: does the software behave as intended? The answer depends on clear descriptions of expected behavior, known as formal specifications. In practice, such specifications are rare because they are difficult to write and maintain. This project studies how to make formal specifications practical, accessible, and central to software development. The project’s novelties are the creation of a large-scale resource that connects everyday language descriptions, precise behavioral rules, and real code; automated methods that derive and update these rules as software evolves; and tools that use them to guide intelligent AI-based coding systems. The project’s broader significance and importance are that reliable specifications enable safer AI-based autonomous systems, reduce costly failures, and increase public trust in critical software infrastructure, including applications in healthcare and finance. The project establishes a repository-level dataset linking natural language documentation, formal behavioral specifications, and code artifacts, accompanied by an automated evaluation pipeline for reproducible assessment of specification accuracy. It develops methods to infer specifications from documentation and program structure, propagate them across code dependencies, and maintain them under version changes to preserve semantic alignment. The project also advances specification-driven development tools that use these formal artifacts as semantic constraints to guide AI-based code generation and to detect behavioral inconsistencies early in the lifecycle. By integrating data infrastructure, inference algorithms, and alignment-oriented tooling, the research strengthens the theoretical and practical foundations of specification-centered software engineering. Its impact extends beyond computer science by enabling more dependable digital systems that society increasingly depends upon. 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.

NIH Research Projects · FY 2026 · 2026-05

Calcium/calmodulin-dependent protein kinase II (CaMKII) is a critical serine/threonine kinase that translates cal- cium signals into long-term cellular responses across diverse systems, including the brain, heart, and reproduc- tive cells. This project focuses on CaMKIIγ, the variant essential for egg activation in mammals, seeking to define how its activation and autophosphorylation are regulated during oocyte maturation, fertilization, and early em- bryonic development. Our long-term goal is to unravel how divalent cations homeostasis and their targets, in- cluding CaMKII, are regulated and function in oocytes, eggs, and during embryo development. The overall ob- jective here is to monitor CaMKII kinase activity in real-time using FRET sensors during maturation, fertilization, and parthenogenetic stimulation and elucidate its regulation by autophosphorylation. Our central hypothesis is that the enhanced sensors will accurately report CaMKII activity, enabling us to monitor its activity and dynamics in real-time during oocyte maturation, throughout fertilization, and early cleavages. We further hypothesize that CaMKIIγ T287 phosphorylation is essential to maintaining kinase activity during the extended intervals between Ca2+ rises of fertilization. These hypotheses are supported by preliminary data showing that: 1) the origi- nal CaMKII biosensor FRESCA-1 only partially tracked CaMKII activity in cells in real-time, 2) newer FRESCA versions (FRESCA-2 and FRESCA-3) offer increased sensitivity and dynamic range, and signal stability for more than 6 hours, 3) eggs of Camk2g knockout (KO) mice have no detectable CaMKII activity, and 4) CaMKII activity and Ca²⁺-dependent egg activation can be restored in KO eggs by expressing CaMKII variants. The rationale for our study is that using the newly developed, robust, and sensitive FRESCA biosensors will enable the detection of CaMKII activity in oocytes and embryonic stages, where this has not been possible. This ap- proach will define the functional significance of CaMKIIγ, the sole isoform expressed in eggs, its regulation, and its role during fertilization and development. These insights will clarify the Ca²⁺ signals and CaMKII activity thresholds required for egg activation and their influence on developmental potential and early embryogenesis. Ultimately, it may inform novel fertility treatments and advance our understanding of CaMKII regulation. We will test these questions using the following specific aims: 1) Define the timing and activity of CaMKII during matu- ration, egg activation, and preimplantation embryo development. 2) Determine the role of CaMKIIγ autophos- phorylation in egg activation. This project investigates how CaMKII, a single kinase, can decode vastly different calcium signals—from rapid neural firing to the slow calcium waves of fertilization—by sustaining its activity through autophosphorylation. While this mechanism is well-established in neurons for memory formation, its role in slow-timescale events, such as in mammalian fertilization, remains unknown. By directly visualizing CaMKII activation in real-time, this work addresses a fundamental gap in our understanding of how a conserved regula- tory switch enables signal decoding across diverse physiological contexts.

NIH Research Projects · FY 2026 · 2026-05

PROJECT SUMMARY Faithful DNA replication is essential for genome integrity and successful embryonic development. During the first mitotic division of the mammalian zygote, genome duplication is exclusively dependent on maternal replication factors deposited during oogenesis. The MCM2-7 helicase complex plays a central role in DNA replication licensing, but the function of maternal MCM components during the oocyte-to-zygote transition remains untested in vivo. We have generated the first oocyte-specific conditional knockout (cKO) of MCM4, a core component of the MCM2-7 complex. Strikingly, Mcm4 cKO zygotes (Mcm4 cKO eggs fertilized by wild-type sperm) fail to incorporate EdU, indicating a complete absence of DNA replication, yet over half of them still enter mitosis and cleave into 2-cell embryos, displaying fragmentation and abnormal chromatin structures. This suggests that the mitotic program can proceed without genome duplication, raising critical questions about cell cycle regulation, checkpoint bypass, and chromatin remodeling under replication-deficient conditions. This project aims to define how maternal MCM4 governs zygotic replication and mitotic fidelity and to uncover the epigenetic and transcriptional consequences resulting from loss of maternal MCM4. We will use live imaging, immunofluorescence, and RNA-seq to investigate mitotic progression, chromatin dynamics, and zygotic genome activation in Mcm4 cKO embryos. These studies will fill a major gap in our understanding of maternal control during early development and may offer new insights into the origins of developmental arrest, infertility, and early embryonic loss.

NSF Awards · FY 2026 · 2026-05

Biofilms are thin layers of bacteria that can stick to wet surfaces. They can cause serious problems in hospitals, water systems, ships, and food production. Current methods to stop biofilm growth can involve toxic chemicals, special coatings, or intensive scrubbing. These methods are costly, short-lived, and detrimental to the environment. This CAREER project will take a new approach. It will fabricate surfaces that emit UV light from within the material itself, so that bacteria will not be able to attach to the surface in the first place. The research team at the University of Massachusetts Amherst will use a novel UV-emitting glass to observe exactly how bacteria respond to light at a surface. These findings will guide the expansion of this technology to more complex surface shapes, such as ship hulls, medical devices, and water pipes. The project will also train the next generation of engineers, from K-12 students to graduate researchers, to advance light-based solutions to real-world problems. This CAREER project will employ UV-emitting glass (UEG) technology, optical waveguides that scatter ultraviolet radiation uniformly across a transparent substrate via silica nanoparticles embedded at the core-cladding interface, as an experimental platform to resolve fundamental questions about light-microorganism interactions at the biofilm attachment interface. Thrust 1 will integrate UEG substrates with custom microfluidic flow cells and high-resolution phase-contrast microscopy to quantify dose-dependent cellular kinetics across irradiance levels of 0.1-50 µW/cm², capturing attachment probability, extracellular polymeric substance (EPS) excretion timing, motility transitions, and inactivation thresholds as functions of irradiance and residence time. Parallel optical coherence tomography (OCT) imaging of UEG-integrated flow cells will provide the first in situ, real-time quantification of biofilm structural dynamics, including thickness, porosity, and roughness coefficient, under photoinduced stress, with direct comparison of internal versus external UV delivery modes across systematically varied environmental conditions (pH, temperature, salinity, flow rate, and nutrient concentration). Automated image analysis and machine learning-assisted cell tracking algorithms will extract behavioral transition kinetics and inform the development of predictive mechanistic models relating irradiance, dose, and environmental parameters to cellular and community-level outcomes. Thrust 2 will translate the light distribution principles established through UEG research to scale self-emitting UV surface technology to complex and non-planar geometries, dramatically expanding the range of surfaces and sectors that can benefit from chemical-free, light-driven biofilm prevention. Together, these research thrusts will advance fundamental photobiology and microbial photophysics while generating scalable biotechnology platforms with transformative potential for biofilm prevention in water treatment, healthcare, and marine applications This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NIH Research Projects · FY 2026 · 2026-04

PROJECT SUMMARY The ability to adaptively and appropriately respond to changes in the environment is a crucial behavior. It is cognitive flexibility that enables this adaptive responding. An inability to adaptively shift behavior (cognitive inflexibility) despite negative outcomes is observed in various psychiatric disorders and is a key diagnostic criterion of alcohol use disorder (AUD). The long-term objective of this work is to elucidate how chronic alcohol and stress promote AUD through cognitive inflexibility. In humans, AUD is accompanied by aberrant frontal cortex function alongside deficits in cognitive flexibility. A significant body of work, including that from our lab, has demonstrated impaired cognitive flexibility in animal models following chronic alcohol which can be exacerbated by stress exposure. However, the frontal cortical neuroadaptations that mediate these alcohol and stress induced cognitive impairments remain unresolved. This is a critical gap given that stress is a key risk factor in perseverative relapse for individuals with AUD. My central hypothesis is that functional network changes across the dorsomedial frontal cortex underlie cognitive flexibility impairments following repeated alcohol and stress exposure. Moreover, I hypothesize that changes in modulatory norepinephrine inputs to this region are critical contributors to these flexibility deficits. To test these hypotheses, I will investigate the mechanisms underlying the negative impact of chronic alcohol and stress on cognitive flexibility using a novel attentional set shifting task. In Aim 1 I will use in vivo calcium imaging to investigate alcohol- and stress- associated changes in dorsomedial frontal cortex dynamics underlying behavioral impairment. In Aim 2, I will determine whether chronic alcohol and stress functionally disrupts noradrenergic regulation of the dorsomedial frontal cortex using in situ hybridization readouts and pharmacological interventions to recover cognition. These proposed studies will provide ample training in cutting-edge techniques and facilitate my professional development in the alcohol research field. At the same time, this work will advance our understanding of the impact of long-term alcohol and stress exposure on cortical function and identify potential mechanisms for therapeutic interventions.

NSF Awards · FY 2026 · 2026-04

The Symposium on the Theory of Computing (STOC) is one of the premier annual research conferences that cover the breadth of theoretical computer science. It is a conference of very long standing that has continued to play a formative role in the field; it is also at the leading edge of addressing algorithmic challenges. This project aims to increase the impact of this conference on students and postdoctoral researchers by encouraging and enabling their participation, especially in cases where travel expenses and conference fees would otherwise preclude their attendance. Concretely, this project assists US-based students and postdoctoral fellows in attending the 2026 Annual STOC conference, sponsored by the ACM. The coming STOC will take place in Salt Lake City, during June 22-26, 2026. The project plans to support fifteen to twenty students and/or postdoctoral fellows in attending this important conference. Supporting the development of these early researchers assists them in making significant contributions to the computer science community and society at large. 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.

NIH Research Projects · FY 2026 · 2026-04

Project Summary/Abstract Antimicrobial Polymers for Treatment of Multidrug-Resistant Wound Biofilm Infections The goal of the proposed research is to create a new polymer therapeutic for multidrug resistant (MDR) biofilm infections. The refractory nature of biofilm infections makes them non-responsive to standard antibiotics, a situation exacerbated by acquired antibacterial resistance. In our research, we have integrated the nanomedicine capabilities of Rotello with the wound biofilm expertise of Patel to develop poly(oxanorborneneimide) antimicrobial polymers (PONI-AMPs) These polymers kill biofilm-based bacteria with minimal effects on host cells. PONI-AMP therapeutics had good efficacy (killing ≥99% of bacteria in biofilms) in an in vivo wound biofilm model developed by Patel. In our proposed research, Rotello will 'evolve' PONI-AMP therapeutics using amino acid sidechains as an initial source of molecular diversity. These polymers will be modeled by Van Lehn using atomistic simulations to generate descriptors that will be used in conjunction with biological properties (antimicrobial efficacy and mammalian cell toxicity) to provide an integrated synthetic, computational, and machine-learning feedback strategy for designing new polymers. The polymers will then be incorporated into a hydrogel to provide controlled release of PONI-AMP to treat wound infections. PONI-AMPs in hydrogel will be tested in vitro and in vivo using realistic and challenging wound biofilm models. Aim 1. Maximize antibacterial and antibiofilm activity through polymer evolution. Rotello will synthesize polymers featuring diverse sidechain functionality and screen for antibacterial and antibiofilm activity against multiple MDR pathogens, with testing of mammalian cell viability performed to determine therapeutic selectivity. Van Lehn will integrate computational chemistry with machine learning to develop descriptors that will be used to evolve polymers to maximize antimicrobial efficacy and minimize mammalian toxicity. Broad-spectrum activity will be targeted, including methicillin-resistant Staphylococcus aureus (MRSA). Rotello will screen evolved PONI-AMPs against MRSA and other pathogens, and Patel will perform testing in multi-pathogen models. Aim 2. Wound hydrogels for nanoparticle delivery. Rotello will develop hydrogel materials (PONI-AMP-hydrogel) for wound covering and PONI-AMP delivery that will provide intimate and sustained contact with wounds and controlled delivery of polymers to biofilms in wound beds. Rotello will screen for efficacy using bioluminescent MRSA biofilms, and Patel will determine efficacy against multi-pathogen biofilms. Aim 3. In vivo and ex vivo testing of PONI-AMP-hydrogel antimicrobials. Patel and Rotello will perform in vivo murine studies of PONI-AMP-hydrogel treatments using collaboratively developed bioluminescent MRSA biofilm wound models, focusing on bactericidal activity, wound healing, and purulence reduction. Patel will perform multi-pathogen murine studies and ex vivo studies using multi-pathogen infected porcine dermal explants.

NIH Research Projects · FY 2026 · 2026-02

Project Summary Chikungunya (CHIKV) is an RNA alphavirus that infects 3 million people in 45 countries including the US annually. Acute infection is flu-like, but in 40% of infections, debilitating joint pain emerges that can last for years. Infection during pregnancy also results in severe encephalopathy in newborns or aborted fetuses. Viral proteases are effective antiviral drug targets and are the standard of care for viral diseases (e.g HIV, hepatitis C, SARS-CoV- 2). The activity of the nsP2 protease from CHIKV (CHIKVP) is vital for infection. Inhibition of CHIKVP blocks processing of the viral polyprotein, prevents viral replication, lowers viral titers and stops disease progression. Thus, CHIKVP is an excellent antiviral drug target. To date, no effective antivirals of CHIKVP have been approved for acute or chronic infection. Our ultimate goal is to use insights into CHIKVP structure and dynamics to develop an inhibitor to oppose CHIKV infection, the resulting chronic pain and prevent pediatric neurological syndromes. CHIKVP is composed of a protease domain and a methyltransferase-like domain (MTL). To date, no functions of the MTL have been identified. In a search for novel CHIKVP binders, we identified ligands that bind to the MTL at an elongated cavity and allosterically inactivate the protease. The site shares structural homology with S-adenosyl methionine (SAM) cofactor binding sites, but does not bind SAM. The allosteric site binds to GTP, which suggests that a function such as RNA binding may be conserved in the MTL. Here we propose a research strategy for the development and direct comparison of CHIKVP active-site and allosteric inhibitors. We will build compounds derived from a large compound screen and also build from MTL-binding fragments we have already identified. We have developed NMR approaches that allow us to readily distinguish active-site from allosteric inhibitors. Importantly, we have developed approaches that allow us to monitor activity, binding and dynamics in solution without having to rely on freezing samples which is required for other structural techniques, to inform our inhibitor design. Recent data have suggested that RNA plays a critical role in CHIKVP function, enhancing protease activity. We have identified a site that we hypothesize binds RNA and describe a series of studies to understand the mechanism by which RNA impacts protease function. We will bring all these structural insights into our inhibitor development approach. At each step of development, we will closely monitor efficacy against viral infection for CHIKV and other related alphaviruses to determine whether pan-alphaviral inhibition is achievable with a given class of compounds. Critically, we will also implement a directed evolution approach across both domains of CHIKVP to predict the susceptibility of our inhibitors to resistance mutations. This will enable us to develop enduring antivirals and will also address longstanding unanswered questions about the favorability of allosteric inhibition in antiviral drug development.

NSF Awards · FY 2026 · 2026-02

This award supports the DAWN-IRES Scholars Program, an international summer research experience for U.S. undergraduate and graduate students at the Cosmic Dawn Center of Excellence (DAWN) in Copenhagen, Denmark. DAWN specializes in studying the formation of the first stars, galaxies, and black holes - addressing some of the most profound questions in modern astronomy. Through a 10-week immersive program, students engage in globally competitive astrophysics research while developing critical skills in data analysis, communication, and cross-cultural collaboration. The program is designed to broadly encourage applications from students across a wide range of backgrounds. In addition to one-on-one research mentoring, the student cohort participates in community-building activities, professional development workshops, and public research presentations in both Denmark and the United States. The program promotes global collaboration, scientific discovery, and a more diverse STEM workforce. This project investigates the physical processes that governed the formation and evolution of the earliest galaxies, with a focus on the epoch of cosmic dawn. Working alongside faculty at the Cosmic Dawn Center and U.S.-based collaborators, students contribute to projects that explore questions such as: What caused cosmic reionization? How do the first galaxies assemble and shut down star formation? Research methods span observational astronomy - using data from a premier suite of space- and ground-based telescopes (including the James Webb Space Telescope, the Hubble Space Telescope, and the Atacama Large Millimeter/submillimeter Array) - as well as theoretical modeling enabled by state-of-the-art hydrodynamical cosmological simulations and advanced data science techniques. Participants gain experience in scientific computing, proposal writing, and independent, data-driven inquiry. The broader impacts of the program include training the next generation of astronomers, building the STEM workforce and strengthening U.S. leadership in global astrophysical research through sustained international partnerships. 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

Collaborative Research: Tectonic Influence on the Greenland Ice Sheet (TIGRIS) The evolution of the Greenland Ice Sheet (GrIS) strongly depends on its underlying geologic structure. Changes to the ice sheet can cause the Earth’s crust and mantle to deform, with the amount of deformation being controlled by variations in the stiffness and thickness of these geologic layers. How the solid-Earth responds can either hinder or enhance ice loss, and this ice-Earth feedback mechanism plays a critical role in determining GrIS behavior. This project aims to evaluate the stability of the GrIS under different environmental conditions by employing an advanced computer model that combines ice-sheet, atmospheric, and geologic constraints. Results from this work will inform estimates of both past and future global sea-level change. Subglacial solid-Earth parameters are largely based on geophysical observations; however, conflicting interpretations of the geologic structure beneath Greenland limit our understanding of GrIS stability. Key portions of Greenland have been under-sampled, and prior studies have often only utilized data from select seismic networks. This project will develop new, self-consistent models of the solid-Earth structure beneath Greenland by combining geophysical observations from multiple networks with those from a new seismic deployment in central-eastern Greenland. Those new solid-Earth constraints will then be incorporated into a state-of-the-art, fully coupled tectonic-atmospheric-cryospheric modeling framework to evaluate the critical thresholds for ice-sheet recovery under different environmental scenarios. Three fundamental hypotheses will be tested: (a) solid-Earth structure plays a first-order role in the long-term future evolution of the GrIS as well as its response to past warming and cooling episodes; (b) under certain projected future warming scenarios, the GrIS will not fully retreat given feedbacks that are controlled by the solid-Earth structure; and (c) the interplay between different feedback mechanisms will result in at least partial ice-sheet recovery, and the GrIS will be resilient in the long term (10-20 kyrs). 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 · 2025-12

Corrosion is a major contributor to structural deterioration in steel structures, leading to increased maintenance costs and reduced reliability. Across the United States, tens of thousands of steel bridges are in poor or fair condition, primarily due to corrosion-related damage. This research project investigates the use of cold spray metal additive manufacturing as a method to restore corroded steel bridge components directly on-site. Unlike traditional repair techniques that require extensive welding, cutting, or complete replacement, cold spray offers the ability to apply new metal onto corroded areas without melting, allowing for localized repairs with minimal surface preparation. Through full-scale testing on real corroded bridge beams, the research will evaluate whether this technique can restore the mechanical properties of damaged steel members to levels suitable for structural use. The approach could minimize waste and lead contamination while enabling targeted repairs using portable, field-deployable equipment. By integrating advanced scanning, quality control, and durability analysis, this work intends to support safer and longer-lasting infrastructure. The work also includes development of inspection protocols to assess repair quality. Additionally, the project provides valuable educational and workforce training opportunities in manufacturing and civil engineering, contributing to the resilience of transportation networks. The research investigates the mechanical feasibility, scalability, and field-deployability of cold spray additive manufacturing (CSAM) for repairing corroded steel structures. Key objectives include: (1) achieving deposition with mechanical properties comparable to structural-grade steel, (2) validating portable cold spray equipment on full-scale, naturally corroded bridge beams, and (3) integrating lead-removal, waste-capture systems, and a novel durability life analysis into the repair process. The methodology combines extended mechanical testing of composite steel coupons, optimization of cold spray deposition parameters, and development of non-destructive evaluation techniques. Digital workflows will be developed looking to to profile corrosion using 3D scanning and apply targeted repairs. Repaired bridge beams will undergo full-scale mechanical testing to validate restoration of load-carrying capacity. Outcomes look to include validated protocols for field application of CSAM in infrastructure repair, new modeling frameworks for corrosion mitigation, and contributions to the fundamental understanding of solid-state additive repair systems for civil infrastructure. 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

Traditional machine learning often involves collecting data from multiple sources, which can raise significant privacy concerns. One approach has emerged as a promising solution to solve this challenge by enabling models to be trained across many different sources without directly sharing private data. This approach has become particularly valuable in sensitive sectors such as medical diagnostics, where individual data privacy is legally protected. Despite these advancements, existing systems for training models across multiple sources lack standardized assessment tools, posing challenges to research reproducibility, validation, and trust. Without proper testing tools, organizations cannot verify that their privacy protections work as intended, creating barriers to adoption in critical areas like healthcare, finance, and national security. This project addresses this challenge by developing comprehensive testing tools that ensure privacy-preserving artificial intelligence systems work reliably, serving the national interest by enabling secure collaboration on AI development while protecting individual privacy, supporting American competitiveness in artificial intelligence technologies, and strengthening data security across critical infrastructure. This project designs, develops, and sustains FLTest, an interdisciplinary testbed that automates privacy and robustness evaluations in federated learning systems, addressing gaps often overlooked by traditional tools. The research activities include developing automated test orchestration frameworks, implementing privacy attack simulation models, creating configuration vulnerability detection systems, and building recommendation engines for optimization. The testbed's key innovation streamlines evaluations through automated orchestration assisted by a pitfall checker that detects configuration issues and vulnerabilities in privacy evaluations. FLTest empowers both novice and expert users with actionable insights tailored to real-world applications. The team will validate FLTest across multiple domains and datasets, develop standardized benchmarks for assessment, and create detailed reporting mechanisms for security analysis. By utilizing distinct datasets and offering a standardized solution, FLTest verifies model privacy and robustness across heterogeneous data distributions, supporting the development of reliable privacy-preserving federated learning systems. The project includes collaboration with three industry partners to ensure practical adoption and long-term sustainability. 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

Quantum networks promise to revolutionize technology by enabling provably secure communication, ultra-precise sensors for scientific discovery, and new forms of distributed and blind quantum computing. However, current quantum hardware is noisy, error-prone, and inefficient, making the reliable distribution of entanglement -- the key resource powering these applications -- a formidable challenge. This project seeks to overcome these hurdles by developing a novel, comprehensive framework to design and operate quantum networks with maximum efficiency. By creating intelligent control policies that are co-designed with the underlying hardware, this work will enable near-term quantum systems to perform tasks that are currently out of reach. This research serves the national interest by accelerating the development of a secure quantum communication infrastructure, a cornerstone for national defense and economic prosperity. The project’s advancements will also support scientific progress by enabling new instruments, such as quantum-enhanced telescopes, and will foster economic welfare through applications in drug discovery and materials science. The project will produce open-source software tools, benefiting the entire research community. Furthermore, it includes significant educational components, developing new university curricula, supporting undergraduate research, and creating pathways for students to enter the quantum information science workforce. This project's central goal is to develop a full-stack framework for the design and optimization of control policies for distributed quantum systems. The approach integrates analytical modeling, high-fidelity simulation, and machine learning to discover robust, hardware-aware protocols. The methodology begins with analytically inspired policy ansatzes that are then refined using Reinforcement Learning (RL) within a realistic simulation environment that captures complex hardware characteristics. To overcome the high computational cost of training, the framework will leverage a combination of high-fidelity simulations for validation and newly developed, efficient surrogate models for both quantum state dynamics and network-level control logic. Key technical innovations include moving beyond simplistic noise models to more faithful representations of biased Pauli noise, amplitude loss, and coherent errors. The project will develop application-specific utility functions for key use cases, including quantum key distribution, quantum-enhanced interferometry, and blind quantum computing, enabling optimization for true performance rather than proxy metrics. The research will explicitly address critical challenges in applying RL to quantum networks, such as state-space explosion, sparse and non-additive rewards, and the generation of deployable policies that operate with local knowledge. The project's outcomes will be a set of optimized entanglement distribution protocols and a principled, scalable methodology for evaluating quantum network utility. 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

Next-generation 6G wireless communication systems are anticipated to bring ultra-high speed, ultra-high reliability, and ultra-low delay global coverage. For wireless communication technologies deployed to date, the radio-frequency (RF) signal propagation environment has been treated as a fixed constraint and limitation, under which theories are developed, standards are written, and devices are designed and developed. Recently, reconfigurable intelligent surface (RIS) has emerged as an enabling technology to overcome these limitations by proactively controlling and optimizing the RF signal propagation environment for future 6G wireless communications. Installed on building facades or billboards, these large planar electromagnetic RIS structures can redirect signals between wireless access points and user devices, overcoming adverse effects caused by signal blockage. They help improve coverage, suppress interference, and enhance security. However, the state-of-the-art RIS designs still face the challenge of providing true full-space coverage and leave some areas with poor signal quality or no signal at all. This project aims to realize a novel RIS technology capable of achieving true full-space coverage by transmitting, reflecting, and scattering electromagnetic waves carrying wireless signals to any possible direction in 3-D space. In addition, simultaneous signal reception and wireless power transfer will be studied. The success of this project will bring the envisioned next-generation wireless communications closer to reality by enabling wireless signal propagation control at will. The research outcome of this project will be incorporated in undergraduate and graduate classroom teaching to train future RF engineers in wireless technologies. Through summer outreach programs, local high-school students will learn about basic electromagnetic wave propagation principles using similar sound-wave control experiments. This project will investigate a new RIS technology at microwave frequencies for full-space coverage with high power-conversion efficiencies. Reflective RIS as well as simultaneously-transmitting-and-reflecting (STAR) RIS will be designed, fabricated, and experimentally characterized. Specifically, the research will investigate redirecting incident waves into exact endfire directions as well as anomalous directions by a planar RIS to achieve true full-space coverage. Endfire-direction beam-scattering synthesis and enhanced RIS capabilities such as converting propagating wave to surface wave and near-field focusing will be achieved by controlling both propagating wave and evanescent wave. The project will develop RIS prototypes comprising reactively loaded arrays of patch antennas for reflective RIS and stacked planar dipole antennas for STAR RIS. Based on the antenna vector effective height extended to arrays and their linear network treatment, numerically efficient design techniques will be developed and applied to electrically large RIS for practical applications. The project will study design trade-offs between wave conversion efficiencies and array element spacing to cover extreme-angle anomalous reflection and refraction as well as endfire-direction beam scattering. For each of the reflective and STAR RIS types, the project will first build and study static-response prototypes using printed-circuit technologies and a parallel-plate waveguide simulating 2-D wave propagation environment. Subsequently, reconfigurable RIS prototypes incorporating semiconductor diodes and digital control circuitry will be designed, fabricated, and evaluated. Based on the results of the 2-D experiments, planar RIS prototypes of reflective and STAR types will be built to demonstrate full 3-D space coverage. As the proposed technique for full-space coverage is not limited to electromagnetic waves, the same design philosophy for high-efficiency, full-space coverage can be extended to other types of wave physics such as acoustic waves. 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 world is becoming more connected through smart devices such as fitness trackers, environmental sensors, and portable gadgets. Finding reliable ways to power them is becoming more important than ever. Today’s batteries do not last forever, can be unpredictable, and often harm the environment when they are thrown away. Other energy sources, like solar or wind, do not always work depending on the weather or location. To tackle these issues, this project aims to develop a new kind of technology termed Air-gen that can generate electricity from something all around us: the moisture in the air. Moisture in the air can carry electricity, as demonstrated by the lightning we see during a thunderstorm. This means that power can be retrieved anytime, anywhere without relying on the sun, wind, or even a battery. Moreover, this research project seeks to make the Air-gen from natural, eco-friendly materials created by microorganisms, using methods that are scalable and compatible with current manufacturing technology. The research also creates educational opportunities through collaborations with a community college and technology translation programs, aiming to engage high school students, college students, and the local community to promote the importance of science and technology. The intellectual merit of this project lies primarily in its conceptual and technological breakthroughs toward a novel form of sustainable energy, with the potential to significantly advance the vision of ubiquitous powering and computing. The broader impacts of this project include applying the developed technologies to clinical settings such as aging and Alzheimer’s care, promoting technology transfer for commercialization, and creating education opportunities to help enhance public awareness of the critical role that science- and technology-driven innovation plays in the development of sustainable solutions. The goal of this research project is to develop wafer-scale, modular integration of Air-gen technology as a universal and ubiquitous powering solution for Internet of Things and wearable devices. The Air-gen is made from a nanoporous thin film with one surface sealed and the other exposed to the air. The asymmetric structure enables charge separation between the two interfaces, driven by interactions with air water molecules in ambient air, thereby generating electricity. To achieve this overarching goal, the project will engineer microorganisms to serve as sustainable bio-factories for producing protein nanowires as the core material used in Air-gen fabrication. Wafer-compatible and scalable integration methods are proposed and investigated to scale up Air-gen power production. These devices are implemented in sleep and environmental monitoring systems to demonstrate their potential as a reliable, self-sustaining power source. This research leverages multidisciplinary expertise, including microbial material engineering, electronic device fabrication, and circuit and system integration. The continued development of Air-gen technology can enable next-generation power solutions for a wide range of systems, including wearable electronics, mobile computing, biomedical devices, and autonomous microsystems. 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

Cooperation of multiple devices to learn and make decisions based on their environment is especially valuable for Internet of Things (IoT) and other applications. Existing algorithms for cooperative learning often assume that all devices face the same set of decision choices, which is not often the case in networked system settings. This project, CoLeNe (Cooperative Learning in heterogeneous Networks), proposes to design and evaluate algorithms for multiple devices to cooperatively learn for decision making over a large set of choices in computer networks as the agents may face different sets of decisions. The developed algorithms allow devices to collaboratively explore the decision space and identify the best option from a large set faster. The project also applies these algorithms to the decision-making cases in networks and demonstrates their usefulness through the examples. In addition, the research effort is paired with educational and outreach initiatives that introduce students to the theory and practice of cooperative learning in networks. This proposal aims to develop cooperative online learning algorithms that are communication-efficient and robust to heterogeneity in computation, data, and privacy across agents. It consists of two research thrusts: (1) theoretical foundations and algorithms. The project will extend existing theoretical work on online learning to the settings of multiple heterogeneous agents with different sets of decision choices and privacy constraints on the information they can share with other agents. It develops algorithms for multiple devices to cooperatively learn to make optimal decisions. (2) Implementing two network applications. The developed cooperative online learning algorithms will be adapted to two applications in edge networks: distributed placement of computing applications across a network of devices, and optimization of wireless network configurations and transmission schedules. The learning framework developed through this project is expected to have broad applicability across IoT as well as other domains involving distributed, heterogeneous learning systems. 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 swift advancement in new energy sources, industrial automation, automotive controls, and space exploration demands the investigation of harsh-environment electronics, often withstanding temperature ranges from hundreds to several thousands of degrees Celsius (°C). For instance, space science communities have advanced electronic components to explore Venus’s atmosphere, which can reach temperatures above 500 °C. Turbine engines in aerospace industries utilize high-temperature sensors for remote pressure transducer interfaces, digitally interconnected actuators, and digital engine controls. The 4th-generation (Generation IV) reactors will function at coolant temperatures higher than light water reactors, reaching around 1000 °C. Hence, the instrumentation and sensors utilized for real-time health monitoring must function in these harsh conditions. Lastly, hypersonic vehicles reach speeds greater than Mach 5, where aerodynamic heating affects air flow, resulting in temperatures exceeding 1000 °C on the vehicle’s surface. As a result, surrounding air molecules ionize and create a buildup of plasma that interferes with electromagnetic waves. Despite its importance, technical challenges exist in radio frequency sensing and communications associated with high-temperature environments. This project will establish new research fields in extreme environments, high-frequency sensing, and high-temperature communications. Because enabling sensors and communication systems in extreme conditions is of paramount importance in designing satellites, spacecraft, automobiles, and space-exploring scientific probes, the proposed methods will benefit future applications including massive global satellite communication networks realized by tens of thousands of Earth-orbiting nano-satellites, hypersonic delivery/transportation infrastructure, CubeSat-based planetary sensing, as well as many other possible commercial and industrial applications. This CAREER project will integrate research and education programs and provide excellent opportunities for high school and college undergraduate/graduate students to engage in STEM research. This CAREER project aims to establish foundational high-frequency electronics for extremely high-temperature sensing and communications beyond 1000 °C, targeting broad industrial applications as well as aerospace and defense applications affected by hypersonic radio blackout. As far back as the 1960s, aerospace communities conducted flight tests to examine radio interference. However, nearly 70 years after the beginning of space exploration, the hypersonic vehicle’s radio interference remains an unsolved problem. In recent years, antennas operating above interference cut-off frequencies and metamaterial-inspired structures have shown new directions, but they lack practical implementations incorporating realistic interference magnitude, effective sensing techniques, system integration and validation, and high-temperature effects. The proposed research rests on the premise that dielectric and ceramic materials exhibit excellent electromagnetic (EM) properties, thermal isolation, and heat tolerance. They are ideal for high-frequency radiating components and for integration with active circuits to build hypersonic transponders and sensors. The project will start with experimental modeling of ceramic materials’ EM properties and thermal expansion in high-temperature conditions and investigate innovative temperature-dependent compensation techniques across extensive temperature ranges, at microwave and millimeter-wave frequencies. The scope and approaches to overcome radio interference challenges include: (1) 3D printed all-ceramic meta-structure, probe, and high-temperature apparatus to compensate for extreme-heat-induced plasma layers, (2) frequency-agile high-temperature/plasma sensing spectroscopic reflectometers with digital compensation for probe’s temperature-dependent EM properties, (3) phase-noise tolerant retrodirective signal trackers with on-chip analog signal processing, (4) artificial-intelligence-assisted temperature-compensating ceramic fiber harnesses. This CAREER project will enable reliable RF signal transmission/reception and signal integrity monitoring with unprecedented dynamic range and sensitivity in continuously and rapidly changing high-temperature environments for hypersonic applications. This CAREER project will also offer new techniques to integrate thermal protection systems and RF electronics, simultaneously achieving thermal isolation and electromagnetic propagation for extremely high-temperature sensing and communications. 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

From online advertising to content delivery networks and recommendation systems, many modern technologies rely on algorithms that must operate in real time without knowing what will happen next. Researchers in computer science, operations research, engineering, and other fields have developed powerful online optimization and learning tools for making effective decisions in the face of uncertainty, by helping systems learn from past outcomes to improve performance over time. However, these methods are vulnerable to attackers aiming to disrupt the system: fake reviews, ad fraud, denial-of-service attacks, and other attacks can corrupt the algorithms' learning process. Researchers have begun developing "corruption-robust" learning algorithms that are more resilient to attacks; however, significant barriers remain in translating these theoretical advances into real-world systems. This project aims to reduce those barriers by designing learning algorithms that are both theoretically sound and practical to implement, enabling more robust decision-making in real-world applications. This work directly supports the national interest by strengthening the resilience of critical cyberinfrastructure and advancing the scientific foundations of trustworthy AI. This project advances the theoretical foundations of online optimization and learning by explicitly incorporating robustness to security threats and adversarial corruptions. While online learning has been extensively studied for decades across computer science, control theory, economics, and operations research, a rigorous understanding of its vulnerabilities and resilience under malicious attacks remains underdeveloped. This project fills this gap by developing new corruption-robust algorithms for sequential decision-making problems, particularly in settings involving combinatorial action spaces and resource constraints in both single-agent and multi-agent learning environments, addressing various classes of adversarial threats. The research involves designing algorithms with provable performance guarantees in the presence of corruption, analyzing their theoretical properties, and developing efficient implementations. To validate its impact, the project evaluates the proposed methods in two real-world applications: probabilistic maximum coverage for content delivery networks and online learning to rank in social and e-commerce platforms. The project's outcomes will contribute broadly to secure and trustworthy decision-making systems. 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.

NIH Research Projects · FY 2025 · 2025-09

PROJECT SUMMARY/ABSTRACT Nearly 20% of school-age children have a communication disorder that limits their ability to participate in daily life, such as classroom discussions or conversations with peers at school. Speech-Language Pathologists are essential for improving the communication of children with communication disorders in schools. Intervening during the school-age developmental period (ages 5-12) is key to leveraging the cognitive, social, and linguistic growth that occurs during this time. Targeting communicative participation (i.e., the child’s ability to communicate while taking part in everyday life) has been documented as an effective approach to support children’s communication and well-being. Directly supporting communicative participation is critical to prioritizing meaningful life participation as a key focus of speech/language intervention. However, less than 1/5 of Speech- Language Pathologists focus their intervention on supporting communicative participation. The research-to- practice gap is clear: researchers have developed a guide that describes how to target communicative participation as an intervention outcome (Baylor & Darling-White, 2020), but this guide is not being used with school-age children with communication disorders. This proposal aims to address this gap by examining the barriers and facilitators influencing Speech-Language Pathologists’ use of communicative participation intervention outcomes (Aim 1) and analyzing Speech-Language Pathologists’ proposed adaptations to the communicative participation intervention guide (Aim 2). By using an implementation science approach, this project directly aligns with the NIDCD Notice of Special Interest in Implementation Science in Communication Disorders (NOT-DC-24-024) and Theme 4 of the NIDCD Strategic Plan (Translate and implement scientific advances into standard clinical care). We will use a convergent mixed methods approach that combines the findings of an online survey (quantitative; Aim 1) and focus groups (qualitative; Aim 2). By partnering with school- based Speech-Language Pathologists, we aim to understand the barriers and facilitators to using these intervention outcomes in schools and adapt the intervention guide based on their direct input. This research will lay the groundwork for future studies focused on reducing the barriers, elevating the facilitators, and implementing the communicative participation intervention guide in school settings with children with communication disorders. Focusing on the Speech-Language Pathologists’ critical role in supporting communicative participation will ultimately benefit the many school-age children currently struggling to express their ideas and communicate in everyday situations at school.

NIH Research Projects · FY 2025 · 2025-09

Project Summary A typical course of antibiotics for tuberculosis (TB) lasts for six months. Subpopulations of non/slow-replicating Mycobacterium tuberculosis have long been hypothesized to be tolerant to antibiotics and contribute to lengthy treatment. Pathways that are active in quiescent M. tuberculosis are high value targets for drug discovery, particularly extracellular components that are freely accessible to small molecules. Cell envelope recycling is an example of a vulnerable pathway in stressed, non/slow-replicating M. tuberculosis. Recycling of the trehalose component of the outer mycomembrane, which occurs in several in vitro models of stress, promotes M. tuberculosis survival in macrophages and in vivo as well as antibiotic tolerance. While the mycomembrane is a distinctive attribute of the Mycobacteriales cell envelope, these organisms also have a more-conserved peptidoglycan cell wall. Very little is known about cell wall turnover and reuse in the Mycobacteriales. In other bacteria, cell wall recycling genes usually do not contribute to fitness in standard laboratory conditions but can support survival in stationary phase and tolerance to certain cell wall-acting antibiotics. Using bioorthogonal metabolic labeling, the investigators serendipitously discovered that M. tuberculosis, M. smegmatis, and Corynebacterium glutamicum recycle the peptidoglycan sugar N-acetylmuramic acid in a manner that is not explained by their known repertoire of cell wall recycling genes. The central hypothesis of this application is that Mycobacteriales MurNAc recycling occurs via a novel mechanism and promotes survival in stressed, non/slow-replicating M. tuberculosis. To test, the investigators will identify and characterize MurNAc recycling genes in M. tuberculosis and determine the phylogenetic distribution of the pathway within and outside of the Mycobacteriales. Successful execution of the proposed work 1) will reveal fundamental mechanistic insight into cell envelope homeostasis not previously defined in standard model organisms and 2) uncover new vulnerabilities in non/slow-replicating M. tuberculosis, including inner- and mycomembrane transporters with good accessibility to small molecules. The results will also form the basis for future investigation of MurNAc recycling in vivo and as a potential antibiotic target.

NSF Awards · FY 2025 · 2025-09

Artificial Intelligence (AI) systems are becoming increasingly capable, yet deploying them to produce reliable, intended outcomes remains difficult. Large language models often fail to follow instructions, AI systems that govern important functions sometimes behave unpredictably, and autonomous systems, such as self-driving vehicles or robots, can act in ways that diverge from user expectations. To improve reliability and utility, future AI systems will need to demonstrate that their goals and behaviors consistently reflect and support the intentions of their human users. The dominant current paradigm for AI alignment relies on learning from human preferences over possible actions or outcomes of an AI system. However, such methods make a number of assumptions about how preferences should be interpreted and ignore many potential sources of error. The aim of this project is to improve the scientific characterization of human preferences in the context of AI alignment and leverage that knowledge to practically improve AI systems. More specifically, Reinforcement Learning from Human Feedback (RLHF) is now at the core of many of the most successful contemporary approaches to AI alignment in applications ranging from robotics to language modeling. RLHF aims to align a policy with the desires implied by human preferences between pairs of trajectories, outcomes, or model outputs. However, such approaches typically rely on very strong assumptions about the meaning of human preferences---for example, that a preference between two trajectories implies that one has a higher utility than another; or that non-utility-related factors do not add noise to preferences, such as the unequal cognitive effort required to evaluate each presented option. If the fundamental interpretation of preferences is flawed, then all of these approaches will also fall short. Thus, the goals of this proposed work are to: (1) Devise and perform human studies to systematically identify the drivers of human preferences in the context of AI alignment -- both core drivers and noise factors; (2) More accurately model core drivers of human preferences to improve the quality of RLHF and AI alignment; (3) Develop data collection best practices to eliminate or mitigate the effects of undesirable noise in human preferences; (4) Demonstrate that these techniques scale to complex problems of practical interest, including self-driving cars and large language models. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

This project makes foundation grantmaking data more accessible and understandable to scientists and the public by building an open access database tracking foundation grants to nonprofit organizations in five U.S. metropolitan areas over time. Philanthropic foundations play a vital role in supporting essential programs in areas such as human services, health, education, and the arts, yet the low amount of research on foundation grantmaking is largely due to limited access to reliable, structured grants data. By showing where foundation dollars go, who benefits, and what issues receive support over time, this project enables interdisciplinary research and advances public understanding of philanthropic behavior and its societal impact. The project addresses a notable gap in the scientific understanding of foundation grantmaking through two activities. First, it curates a Longitudinal Foundation Grants Database (LFGD) by extracting and structuring data from foundation tax filings, focusing on grantmaking in five major U.S. metropolitan areas from 2020 to 2023. While the release of Form 990 makes foundation tax data publicly available, its nested XML format and the complexity of funder-nonprofit grants render it difficult to parse and access for research. This project stands out not only for the scale and longitudinal depth of its dataset, but also for its novel use of large language models (LLMs) to classify unstructured grant descriptions into structured variables such as purpose, issue area, and target population. Second, this project applies both descriptive and inferential network models, respectively, to map and analyze the structure and evolution of funder-nonprofit grants networks over time. Drawing on theoretical frameworks from organizational science, the project investigates how factors like organizational status, organizational attributes, and institutional environments shape grantmaking networks and how foundations respond differently to institutional pressures. The resulting comprehensive open-access database will include detailed grant-level data, LLM-generated insights from grant descriptions, organizational characteristics of foundation funders and nonprofit grantees, and network data capturing relational dynamics in grantmaking over time. This project creates a publicly accessible database that serves as a resource for examining the role foundations play in advancing the public good. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

Understanding how energy moves through nonlinear systems is essential for progress in many areas of science and engineering, including fluid dynamics, neuroscience, and the design of advanced materials. This project studies a mathematical model known as a nonlinear oscillator chain, where interactions between neighboring components can create complex, cascading flows of energy between different scales. Such systems serve as simplified yet powerful representations of more complicated physical processes, such as ocean turbulence or signal propagation in the brain. This project supports fundamental research in probability and applied dynamical systems, as well as the development of new computational tools for analyzing high-dimensional stochastic systems that also inform coupled neuronal oscillators and machine learning algorithms. Through student training activities, this work will help build a capable STEM workforce, contributing to national priorities in scientific advancement and education. Recent breakthroughs have drawn new connections between nonlinear dispersive equations and wave kinetic equations (WKE), with particular interest in understanding how energy cascades through scales in weakly nonlinear regimes. A central object in this theory is the Kolmogorov–Zakharov (KZ) spectrum, a formal steady-state solution of the WKE that reflects how energy transfers across modes. This project investigates a class of nonlinear oscillator chains—called energy cascade systems—that are derived from nonlinear dispersive equations and serve as finite-dimensional approximations to wave turbulence. The principal goal is to rigorously study the nonequilibrium steady states (NESS) of these systems and their connection to the KZ spectrum. Building on recent success of proving the stochastic stability of NESS in short cascade chains using a newly developed Feynman-Kac-Lyapunov method, this work will extend these results to longer chains, addressing a key open problem in the field. Complementing this analytical work, the project will extend the principal investigator's earlier development of deep learning-based Fokker-Planck solver to genuinely high-dimensional systems and to Fokker–Planck eigenfunction problems. The numerical work will support the study of oscillator chains and enable broader applications in coupled neuronal systems and machine learning. These combined efforts will advance the mathematical understanding of energy cascades, nonequilibrium phenomena, and high-dimensional stochastic 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 2025 · 2025-09

Warfighters must maintain agility and performance in extreme conditions such as navigating rugged terrain, carrying heavy loads, and enduring prolonged exertion, often while facing unpredictable threats. Wearable technologies like robotic exoskeletons and advanced footwear have the potential to enhance warfighter performance and reduce injury risk. However, current design methods often rely on one-size-fits-all approaches and fail to account for how individuals adapt to these devices in real-world settings. This project addresses that gap by developing Digital Twins of agile locomotion in the form of personalized, data-driven simulations that model the complex and dynamic interaction between human movement, wearable technology, and the environment. By integrating real-time physiological and biomechanical data, these models enable better design, training, and deployment of active wearable technology to improve human agility. In addition to advancing national defense and security, this work has broad societal benefits to public health as the mathematical modeling techniques developed can also be used to improve wearable technology design for other user populations, such as those with motor impairment. The overarching goal of this project is to develop mathematical methods enabling an advanced Digital Twin model of human agile locomotion, aimed at optimizing the design of advanced footwear technology to enhance human agility and mobility. In order to accomplish this, this project will advance the state of the art in statistical surrogate modeling, which currently is focused on vector-valued parameters, to accommodate parameters which are functions. This will require significant mathematical and methodological innovation as the parameter spaces are thus infinite dimensional. The investigators will develop an approach which searches within a manifold of finite but increasing dimension to find candidate functions to test. This new methodology will be developed using data from human locomotion when using wearable technologies in a lab setting. The investigators will first deploy this methodology to develop a novel active model reduction method which searches for parameter settings which are not accommodated by the current reduced model. Next, they will extend sequential design for data-efficient predictive modeling to the acquisition of functions, enabling a model of human adaptation in response to wearable devices. Finally, they will develop an infinite-dimensional extension of the Active Subspace method for dimension reduction to enable interpretable optimization of wearable devices. Taken together, this work will lead to a general-purpose framework for building Digital Twins of systems parameterized by functions, as well as a specific implementation of a Digital Twin for the complex system of a human bearing wearable technology. 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.