Grants

NSF

This Faculty Early Career Development (CAREER) project aims to enhance the sustainability and resilience of transportation and power systems (TPSs) in response to rapid deployment of electric vehicles (EVs) and clean energy. Traditional, system-specific approaches are often inadequate or unable to address close couplings and decentralized decision-making scheme. This research meets this fundamental challenge by offering a novel mechanism design for system-level planning and operation. The methodologies developed have the potential to extend to other infrastructure systems, where heterogeneous stakeholders interact with each other over a large-scale network. Research findings will help inform future strategies for EV adoption and grid integration of intermittent clean energy sources. The integrated research and education activities are intended to facilitate knowledge transfer to students, practitioners, and the public, including K-12 and college students, utility companies, and transportation planning agencies. The scientific goal of this CAREER project is to advance the understanding of the mechanism design of decentralized TPSs. More specifically, the research efforts will advance the knowledge on (1) network modeling strategies to elucidate the spatiotemporal interactions among heterogeneous and decentralized stakeholders with incomplete information, (2) optimal information sensing and sharing strategies for decentralized TPSs, and (3) equity-aware market mechanism design to optimize TPSs leveraging EVs and clean energy. Meanwhile, it will integrate convexification, decomposition, and variational analysis theories to cope with the computational challenges brought by multi-agent interaction, multi-stage decision-making, and multi-dimensional scenarios for TPSs planning and operation. 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

Neuromuscular disorders affect millions of individuals worldwide, yet efficient tools to accelerate diagnosis and assess therapeutic interventions are currently lacking. Existing methods for evaluating muscle health face significant limitations, including clinical impracticality due to cumbersome procedures, reliance on highly trained personnel, high costs, and safety concerns stemming from associated pain or the use of ionizing radiation. The emergence of electrical impedance myography (EIM) offers a promising avenue for assessing muscle health. EIM is sensitive to changes in muscle structure and composition brought about by a variety of neuromuscular disorders as well as by disuse, producing unique disease signatures that will vary with muscle status. Thus, EIM analysis can provide a method to rapidly, quantitatively, and reliably diagnose and monitor neuromuscular diseases at the bedside, act as a tool to help tailor care for individual patients and streamline and improve clinical drug trials. This CAREER project integrates research with educational outreach by offering students hands-on experience in innovative translational research. This is interlaced with a long-term educational objective of mentoring new generations of students by providing them with experiences in cutting-edge research, developing and implementing activity-based style courses to motivate students’ self-learning in the classroom, encouraging students to choose a science, technology, engineering or math (STEM) degree by participating in research, and assisting undergraduate students in their own translational research efforts. This CAREER project will establish the scientific foundations of future generation EIM tools and enhance diagnostic accuracy by integrating artificial intelligence algorithms with simulation and analytical methods to extract quantitative muscle insights that are currently inaccessible. The tools developed and data collected are expected to lead to a deeper understanding of the role played by muscle electrical properties in EIM, understanding that is needed for the development of new and more accurate EIM tools for evaluating neuromuscular disorders (NMD). Research objectives include (1) developing a physics-informed analytical and simulation framework to model the entangled multicellular architecture underlying tissues and automate the extraction of relevant physical and biological information from EIM data, (2) assessing EIM biological variability in silico, and (3) evaluating the robustness of models generating EIM data. In silico simulations and ex vivo measurements will provide proof of principle to optimally determine the minimal yet sufficiently biophysical relevant mechanisms needed to build robust virtual EIM predictions for healthy and prototypical diseased conditions necessary to interpret EIM outcomes in patients. 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

Red Supergiants (RSGs) are a late stage in the life cycle of stars of 9 – 25 solar masses. During the RSG stage, these stars experience significant mass-loss, and this lost mass enriches the interstellar medium, playing a significant role in the chemical evolution of the Universe. This project will use high resolution imaging from optical interferometry (OI) to study RSGs. High resolution imaging of stellar surfaces produced via OI enables the study of changes on stellar surfaces at unprecedented detail. Results from this project will help determine the role of convection in driving mass-loss in RSGs. This research will be conducted at New Mexico Tech, a Hispanic serving institution. Two PhD and four undergraduate students will be supported (per year) by this project to collect observations, participate in data reduction, and contribute to model development. Outreach initiatives will involve public talks, volunteering at state-level science competitions, and collaboration with StarDate to produce radio shows that communicate the excitement and results of the research. High resolution imaging of stellar surfaces produced via optical interferometry (OI) enables the study of changes on stellar surfaces at unprecedented detail. This project will use the Center for High Angular Resolution Astronomy (CHARA) Array and Magdalena Ridge Observatory Interferometer (MROI) to collect OI data for image reconstruction of RSGs at different stages of evolution. Stars of angular diameter greater than 4 mas will be used to study both large convection features and smaller granulation-like features. By comparing images reconstructed from data covering continuum wavelengths to those covering molecular bands, it is possible to study the link between surface activity and the formation of molecules further up in the stellar atmosphere. The project will also develop new image reconstruction tools to maximize the scientific output of the data. Techniques and models developed in this program will benefit other users of optical interferometers, and all software will be made publicly accessible. The model refinements will have an impact on other aspects of astrophysics, such as the study of stellar evolution in other galaxies. 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

Systemic inflammatory responses are dramatic immune flares that can be triggered by causes such as infection and surgical trauma. These inflammatory responses can be lethal. The goal of this project is to develop a mathematical framework to describe these responses, using a combination of experiments in a model biological system (zebrafish) and novel Artificial Intelligence (AI) and Machine Learning (ML) tools. This project will use methods from genomics to trace the organism’s immune response and develop computational tools to simulate these responses and design interventions to alter their course. This work contributes to biotechnology via development of high-throughput quantitative genomics methods. The project will also develop new computational tools that contribute to the integration of AI with traditional analytical methods in physics and biology. The project has broader impacts on researchers in adjacent fields, including the biomedical research community. The investigators will also engage in educational initiatives that strengthen the future scientific workforce, especially at the expanding interfaces of physics, computational sciences, and biology. Systemic inflammatory responses are dramatic, organism-wide immune flares, characterized by high levels of immune signaling proteins (“cytokine storms”) and infiltration of immune cells into tissues and organs. A unifying feature of these responses is that they are driven by interactions within the immune system itself, as opposed to external stimuli. It remains a major challenge to understand how these apparent runaway responses emerge from interacting components. While immunologists have long discussed immune states, we lack concrete characterizations of the phase space. Thus, many fundamental questions remain open: to what extent are responses canalized onto predictable paths? What variables control key tipping points to runaway responses? The project will combine theory with experiments in larval zebrafish to develop a dynamical systems framework for systemic inflammatory response in this tractable model system. In Aim 1, the investigators will perform high-throughput RNA sequencing-based measurements to map phase portraits of systemic inflammatory responses. In Aim 2, the investigators will use machine learning to model the responses and test interventions in silico. In Aim 3, the investigators will use live imaging zebrafish to trace response dynamics and test model predictions. This project will yield the first measurements of the basins and attractors governing the behavior of systemic inflammatory responses and provide a predictive mathematical framework for altering the course of these responses. This project integrates broader impacts in both research and education. The project will provide proof-of-principle for a dynamical systems framework that enables prediction and manipulation of systemic inflammatory responses. This framework can be applied by researchers in adjacent fields, including the biomedical research community, to human disease, ultimately improving societal well-being. The investigators will also broaden and strengthen participation in the physics major at the collegiate level via two initiatives: developing collaborative learning sections for the University of Chicago Honors Introductory Physics sequence and integrating physics of living systems problems into our first-year curriculum. These initiatives will strengthen retention and training for physics students in our physics major and introduce students to the power of physics for biological 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

Computer simulations, based on physics, can be used to predict how galaxies grow over time. These cosmological simulations connect the growth of dark matter to stellar and gaseous processes. The predictions that these simulations provide are often sensitive to various arbitrary choices which introduce uncertainty that is rarely quantified. This program will develop a novel analysis, based on AI, that will allow for the characterization of uncertainties in cosmological simulations in a novel way. This research will be conducted at the University of Virginia and support a graduate research student who will participate in the model development. Educational initiatives include a Research Experiences for Undergraduates program and an annual schedule of teacher training activities. This project will advance the field of galaxy formation by executing a simulation suite designed for uncertainty quantification across physical models, cosmological assumptions, and resolution scales. A fully automated simulation pipeline will enable the execution of thousands of zoom-in simulations at varying resolution and halo mass, sampling parameters such as supernova feedback strength, AGN efficiency, Omega-matter, and sigma-8. Simulation outputs will be paired with dark matter-only merger trees and analyzed using Graph Neural Networks (GNNs) that learn to predict galaxy properties across the sampled parameter space. The models will be extended to predict spatially resolved galaxy properties using a hybrid of GNNs and diffusion-based generative models, allowing the sampling of entire profile distributions conditioned on both formation history and simulation assumptions. The tools will enable direct comparison across physical models and offer a statistically grounded framework for testing small-scale tensions in Lambda-CDM. 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

This Faculty Early Career Development Program (CAREER) grant supports research that intends to enhance decision making in decentralized systems. Decentralized decision-making naturally arises in many scenarios, notably in supply chains and transportation and logistics, where multiple self-interested agents share a finite collection of common resources and influence each other's behavior. One crucial challenge arising in decentralized systems is the lack of coordination among interdependent decisions, which are made locally but have broader system-wide implications. This research project will approach decentralized decision-making through a game theoretic framework. For systems where the decisions are chosen from a discrete set, this project explores a computational approach that exploits the discrete nature of the individual optimization problems. The models and algorithms developed will be evaluated on a problem involving multiple loosely coordinated agents providing logistic support for post-disaster operations, with the goal to improve resource utilization and coordination among responding organizations. The educational component of the project aims to engage undergraduates and K-12 students in science through a community-building series and a structured portfolio of outreach activities. This project will integrate techniques from integer optimization and algorithmic game theory to establish new theoretical foundations and practical algorithms for resource-sharing games. In these games, multiple decision-makers, each solving their own integer optimization problem, interact while sharing a common set of resources. The methodology will leverage polyhedral characterizations, sparsity structures and decomposition methods beyond optimization, to provide certificates for existence of equilibrium solutions, algorithms for computing exact or approximate equilibria, bounds on equilibria inefficiency, and strategies to reduce such inefficiency. By exploiting polyhedral and combinatorial structures, the project looks to isolate key sources of complexity and identify broad classes of tractable games, both theoretically and computationally. 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

The real world is abundant with visual information, demanding both humans and intelligent systems to navigate and adapt to its complexity efficiently. Human visual processing can be broadly categorized into two types: intuitive processing, which enables rapid, experience-driven decision-making, and deliberate processing, which involves focused and systematic reasoning for complex tasks. Robust visual intelligence, the ability to integrate these types of processing, is fundamental to real-world problem solving and developing common-sense understanding. However, current artificial intelligence (AI) systems lack this level of robustness, as they often rely heavily on language models to compensate for deficiencies in visual architecture. This reliance becomes a critical bottleneck as tasks and scenarios grow more complex, limiting the adaptability and reliability of AI systems in practical applications. The project aims to build a hybrid, vision-centric framework that integrates intuitive and deliberate visual processing to create more robust visual intelligence. This improved capability has the potential to empower research in many other fields. For example, robotics researchers could use the system to help robots see, move and act more intelligently in challenging environments. Similarly, scientists and educators could use it to better understand complex diagrams, making it easier to uncover new insights and improve learning. The project aims to develop a hybrid approach that combines parametric (intuitive) and non-parametric (deliberate) visual processing methods to build robust visual intelligence. The first direction focuses on developing new ways to learn visual representations using techniques like visual self-supervised learning, language guidance, and generative modeling. And the goal for this direction is to advance vision-centric parametric knowledge to form the foundational layer of intuitive understanding. Building on this, the second direction incorporates human-like non-parametric mechanisms, such as visual search and working memory, to enhance deliberate reasoning capabilities. The third direction integrates these two approaches into a unified, hybrid architecture, where a high-level controller is capable of activating either method as needed for task-specific demands. Finally, the system will be tested in real-world, dynamic environments that extend beyond static image datasets, including tasks requiring long-form video analysis and visual-spatial reasoning. By applying the framework to these new challenges, the project aims to produce more adaptable, reliable, and broadly useful methods that expand the possibilities of vision-based applications. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

This NSF CAREER project aims to enhance electric power grid operators' situational awareness, improve dynamic model quality, and enable online controls to ensure secure power system operation with high penetration of inverter-based resources (IBRs). The project will bring transformative changes to the use of measurements for dynamic state estimation, model deficiency diagnosis and calibration, and measurement configuration, thereby enhancing system reliability and security. This will be achieved through innovative dynamic estimation theories and algorithms that leverage the increasing diversity of sensors and communication infrastructure, as well as advancements in robust estimation, uncertainty quantification, optimization, and data analytics. The intellectual merits of the project include i) a generalized, computationally efficient derivative-free observability theory, with observability indices tailored for dynamic systems with black-box models, ii) integration of Bayesian inference with robust estimation to develop novel nonlinear dynamic estimation methods and iii) a scalable Bayesian framework for dynamic parameter estimation and uncertainty quantification. The broader impacts of the project include developing the next generation of robust dynamic estimation paradigms for IBR-dominated power systems, and industry-academia collaborative initiatives to promote industry-driven research, course renovation, and training to equip students (including K-12 students, and those from underrepresented groups across different disciplines and diverse backgrounds) with unique experiences in renewable energy technologies, data analytics and power engineering. The rapid deployment of IBRs, such as solar and wind farms, and battery energy storage is changing the dynamic landscape of electric power grids. Traditional steady-state-based static state estimation, used in current energy management systems, is insufficient for capturing these dynamics in real-time operations. This project addresses the critical need for improved dynamic observability and reliable models for system reliability analysis and decision-making. The research objectives include i) developing a generalized derivative-free state and parameter observability theory for black-box and hybrid dynamic systems, overcoming limitations of linearization-based and Lie-derivative-based theories for IBR-dominated systems, ii) fusing robust statistics with estimation and optimization to create nonlinear dynamic estimation methods capable of addressing black-box IBR models, control mode switches, current limiters, anti-windup constraints, unknown controls, and multi-timescale dynamics and iii) designing observability-informed, scalable parameter estimation and uncertainty quantification algorithms to continuously refine power system dynamic 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

This Faculty Early Career Development (CAREER) award supports fundamental research to understand the processing mechanisms of a novel additive manufacturing (AM) approach for high-performance tungsten alloy fabrications. Tungsten alloys hold significant promise for use in demanding applications, such as aerospace, defense, medical, and nuclear fields, due to their high density and excellent mechanical properties. High-temperature AM, such as powder bed fusion or direct energy deposition, has been used to fabricate tungsten alloy components with short lead time. Nevertheless, tungsten alloys are extremely refractory and have a remarkable ductile-to-brittle transition that makes the materials highly sensitive to temperature changes. The large temperature gradient and huge heating/cooling rates in high-temperature AM pose critical challenges to the fabrication because of the inherent brittleness of tungsten alloys to cracking, leading to the difficulty in creating highly dense components. To address the challenges, this CAREER project takes a unique manufacturing strategy to better tailor and control the formation and densification of as-built components through compression-enabled filament-extrusion AM, binder removal, and hybrid-phase sintering integrating both solid-phase densification and liquid-phase strengthening. If successful, this project will advance the innovation of next-generation, scalable AM processes for high-performance refractory alloy fabrications. The integrated educational and outreach activities will inspire and train students at all levels to acquire multidisciplinary knowledge and skills for future excellence in manufacturing. This CAREER project aims to establish a fundamental understanding of the mechanisms underlying the innovative extrusion-based, sintering-assisted AM of tungsten alloys. First, in-situ rolling-induced particle compaction behaviors during the extrusion of green parts will be discovered. A multi-physics simulation framework coupling discrete element method with computational fluid dynamics will be created to identify the distributions of metal particles and interlayer pores. Then, atomic-scale phase transformation and microstructure evolution during hybrid-phase sintering will be unveiled using molecular dynamics and phase field model, respectively. The established mechanistic understanding will be experimentally validated by measuring phase content, grain morphology/size, and relative density. Finally, time-dependent macroscopic shrinkage, ductile-to-brittle transition temperature, and thermomechanical properties of finished tungsten specimens will be quantified using an integrated approach of constitutive modeling, in-situ measurements within a customized furnace, and thermomechanical testing. The research outcomes can be readily generalized to other metal AM processes, such as ink writing, binder jetting, and aerosol jet printing, that are followed by subsequent debinding and sintering. 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

Proposal Number: 2043803 Principal Investigator: Elaheh Ahmadi, Title: CAREER: A novel Gallium Oxide based transistor for low-waste power conversion applications Institution: University of Michigan Nontechnical Abstract There is an urgent need for new device technologies to efficiently manage and distribute electrical power in the 2000V-20000V voltage range. The current technology, however, can no longer meet the efficiency and reliability requirements for these high-power electronic applications. The proposed ultra-high voltage switch will enable efficient high-power switches in the 2000V-20000V voltage range, which is required in many systems, including distributed grid systems, industrial automation, electric vehicles, and electrical mass transit including high-speed trains. The integrated education plan aims to motivate young students, especially female students and those from the underrepresented groups, to pursue STEM studies and careers by direct participation in the proposed research activities. The tutorials prepared for high school students will be made widely available on PI’s research website and YouTube. The scientific results will be disseminated in the form of articles in technical journals, conference presentations, and university seminars. Technical Abstract The goal of this program is to demonstrate E-mode β-Ga2O3 fin field-effect transistors (FinFETs) with breakdown voltages beyond 3kV, specific on-resistance lower than 1 milli-ohm-cm-square switching efficiency higher than 98% at 15kHz frequency. The scientific goals of this CAREER plan are (i) A thorough investigation and experimentation into the impact of slanted-sidewall and N implantation in the inter-fin areas on electric fields to enhance breakdown voltage. (ii) Fabrication and characterization of FinFETs on both metal-organic chemical vapor deposition (MOCVD)-grown and halide vapor phase epitaxy (HVPE)-grown epi-structure to analyze and compare the impact of growth technique, and material quality on device performance. This study assists in selecting the appropriate growth technique for each process module in the future. (iii) A systematic study of the substrate thickness impact on device characteristics such as Ron. (iv) Development of β-(Al,Ga)2O3 regrowth on the fin sidewalls by plasma-assisted molecular beam epitaxy (PAMBE) and a complete investigation of the impact of regrowth conditions on electron mobility in the channel, interface trap density, and FinFET characteristics. This will be the first demonstration of sidewall regrowth in this material system. (v) Development of robust and reliable Hf(Si)O2 dielectrics by PAMBE as well as investigation and analysis of deposition conditions on dielectric quality (breakdown, dielectric constant, leakage, etc.). This will be the first demonstration of in-situ dielectric deposition in this material system. 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

This Faculty Early Career Development (CAREER) award will support research that attempts to advance a new computational framework that integrates high-fidelity physics and machine learning to transform earthquake engineering for geostructural systems. By reducing the high computational costs typically associated with accurate numerical models, this research intends to enable rapid and reliable seismic response predictions for geostructural systems. These predictive capabilities are critical for designing and safeguarding infrastructure, minimizing economic disruptions, and building more resilient communities. The approach will intends to pave the way for better sensor data integration, enhancing understanding of complex geologic conditions and improving risk-assessment tools for engineers and policymakers. Through complementary educational initiatives, including hands-on activities, online tools, and curricular enhancements, the project seeks to improve scientific machine learning literacy and train a new generation of engineers equipped to apply machine learning responsibly. Ultimately, this CAREER award supports national interest by promoting public safety, advancing the progress of science, and fostering robust workforce development. This research aims to develop a physics-informed machine learning surrogate modeling framework for geostructural systems under seismic loading. Unlike traditional black-box machine learning approaches that require massive datasets and can yield physically unreasonable results, this framework attempts to embed the governing equations, boundary conditions, and constitutive relations directly into the model training process. By penalizing violations of these physical laws instead of relying solely on data, the project looks to substantially reduce the need for large-scale numerical simulations for training and improve interpretability. Specific objectives include (1) developing scalable machine learning algorithms for accurately modeling the dynamic response of geostructural systems to transient ground deformations due to seismic wave propagation, (2) developing optimal approaches to accommodate uncertainties in uncertainty quantification tasks, and (3) validating the framework and assessing its effectiveness using experimental data. This research looks to promote seamless integration of high-fidelity data into site-specific and rupture-to-structure seismic risk assessment workflows, advancing the fidelity of hazard mitigation and performance-based engineering decision tools. It will strives to enable optimal utilization and assimilation of sensing data at a system level to calibrate high-fidelity models and characterize biases and errors. The methodologies developed here seek to enhance state-of-the-art earthquake engineering and resilience strategies for geostructural systems, and to also be applicable to the dynamic response of other systems in engineering mechanics, cyber-physical systems, and earth sciences. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

Modern cyberinfrastructure (CI) powers many aspects of our day-to-day life (such as healthcare, finance, electricity, and communication). And, moving forward, our reliance on this infrastructure will grow even more as new use cases (e.g., augmented/virtual reality, autonomous transportation, and remote surgery) and users (e.g., self-driving cars and IoT devices) enter the technological landscape. To meet the strict security and performance requirements of this evolving landscape, the cloud datacenter networks and systems, which form the backbone of CI, must adapt and allocate their (heterogeneous) resources quickly and efficiently. Doing so, however, demands that (compute-intensive) network management and control decisions are applied per packet at line rate in a fast-and-intelligent way. Unfortunately, the dominant solutions available today are neither fast nor intelligent enough to meet these requirements. This proposal aims to bridge this gap between speed and intelligence by developing a holistic platform that allows datacenter operators to execute per-packet AI-driven decisions directly within the network at line rate. The proposal lays out the research across three progressively connected thrusts: (1) designing novel data-plane architectures for per-packet AI, (2) implementing high-level, declarative frameworks for expressing AI objectives (and models), and, finally, (3) developing a suite of per-packet AI applications to build confidence in the utility of the proposed platform. It is a radically new paradigm that converges multiple disciplines (machine learning, networking, and architecture), thus opening pathways for machine-learning researchers and architects—with their cross-disciplinary knowledge—to work alongside network designers to realize the full potential of per-packet AI. 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

Improving STEM learning for all Americans depends on the preparation and development of science teachers who are equipped to translate research-based instructional strategies into meaningful classroom practice. This CAREER project addresses a crucial need in K-12 science teacher education to respond to local school district needs for high-quality science teaching and the role of teacher education programs, particularly at Historically Black Colleges and Universities (HBCUs) but also more broadly at other institutions that prepare STEM teachers, to develop programs that provide prospective teachers the best opportunity for success as science teachers. Specifically, the project aims to advance science teacher education by applying a pragmatic, iterative approach to developing teacher education program resources and tools that will support the implementation of evidence-based STEM teaching and learning practices in K-12 science classrooms. The project will identify evidence-based STEM teaching and learning practices through a systematic review of K-12 STEM education research and resources. Rather than generate new evidence, the project leverages the evidence that already exists to support educators in adapting and sustaining existing high-quality practices that have already demonstrated positive impacts on students' STEM learning. Through close collaboration with teacher candidates, practicing science teachers, and teacher educators, the project will build capacity for instructional improvement in ways that are responsive to local contexts while aligned with national priorities in science education. The project's long-term goal is to increase the effectiveness and coherence of science teacher preparation programs at HBCUs and more broadly in other teacher preparation program contexts, thereby contributing to future science teacher effectiveness and K-12 students' greater achievement in STEM. This five-year CAREER project involves two primary activities carried out across two phases. Phase 1 involves conducting a systematic review of K-12 STEM teaching and learning research and resources to identify and interpret evidence-based K-12 STEM teaching and learning practices. The findings from the review will result in a preliminary framework of evidence-based K-12 STEM teaching and learning practices to inform science teacher education at HBCUs as well as other institution types. Phase 2 involves Participatory Action Research in which pre-service teachers, in-service science teachers, and teacher preparation faculty work collaboratively to refine the framework and develop an observation tool that will help pre-service teachers and their mentors more effectively prepare and use feedback about their science and STEM teaching. Using a continuous improvement framework, participants will collaboratively implement, adapt, and refine vetted science education practices evidenced in the preliminary framework developed during Phase 1 of the study. Guided by a pragmatic theoretical framework, the project will support iterative Plan-Do-Study-Act cycles that allow educators to identify implementation barriers, evaluate instructional impact, and adjust strategies based on local classroom contexts and realities. The project emphasizes formative assessment, reflective inquiry, and collaborative sense-making as mechanisms to support sustainable instructional change. By grounding science teacher development in well-established evidence while allowing for localized adaptation and continuous feedback, the project will yield adaptable models, tools, and insights that can inform science teacher education nationally. Ultimately, this work supports improving STEM learning for all Americans by equipping educators with the knowledge, flexibility, and confidence to use research-based practices effectively in their own classrooms. The Discovery Research preK-12 program (DRK-12) is an applied research program that seeks to significantly enhance the learning and teaching of science, technology, engineering, and mathematics (STEM) by preK-12 students and teachers. Projects in the DRK-12 program build on fundamental research in STEM education and prior research and development efforts that provide theoretical and empirical justification for funded projects. 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

Detailed cosmological observations suggest invisible substances dominate the energy budget of the Universe and play an outsized role in its growth from the Big Bang to today. Theories put forward to explain the fundamental nature of these “dark” components require time-consuming numerical computations to test their predictions against observations from state-of-the-art telescopes. Recent advances in artificial intelligence and machine learning (AI/ML) dramatically speed up such computation, making it possible to explore efficiently a broader range of plausible theories. This project develops AI/ML-based computational tools enabling novel searches for the physical origin and properties of the invisible elements making up the cosmos. These tools leverage complementarity between astronomical observations and laboratory-based experiments in determining the physics governing the growth of the Universe. In parallel, a pilot program is established to mentor and engage students early-on in research. This program focuses on acquisition of transferable skills broadly applicable to careers in STEM, while providing students with a support community informing their sense of relevance within physics. The project aims to determine whether neutrinos have nonstandard interactions that drive cosmic expansion and the growth of large-scale structure. To do so, this project incorporates the latest advances in AI/ML to develop a fully differentiable cosmological pipeline that dramatically speeds up searches for new physics with cosmological data. This approach enables the exploration of a much broader range of cosmological scenarios than previously possible. It also provides efficient sensitivity forecasts for cosmic microwave background and large-scale structure data from CMB-S4 and Rubin Observatory, probing potential synergies among the next decade’s major experimental progress in particle physics and cosmology. The approach enables complementarity studies between ground-based neutrino experiments and observations. In parallel, a new seminar, focusing on important skills, includes software development and AI/ML, time management, written and oral communication, and career planning will be established. 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

Real-world systems are composed of interconnected entities that collaborate to perform diverse yet interrelated tasks, often requiring sequential decision-making. Over the past decade, multi-task learning has emerged as a powerful paradigm for collaborative learning, significantly enhancing efficiency while enabling privacy-preserving knowledge sharing. One of the most promising approaches to learning-based control is dynamic sequential learning, such as reinforcement learning, which learns through interactions with the environment. However, reinforcement learning faces three critical challenges in real-world dynamical systems: data scarcity and heterogeneity, scalability and communication efficiency, and safety. Moreover, achieving provable guarantees in joint learning often requires leveraging underlying problem structures. This CAREER project will develop a unified approach to multi-task representation learning by leveraging the shared representations to offer a viable solution to these challenges, enabling privacy-preserved joint learning in dynamic environments. Research and education will be synergistically integrated to train students in the interdisciplinary field of data science and control, addressing the pressing need for skilled workforce development in this emerging area of societal importance. The central objective of this project is to develop provable methods for multi-task representation learning in bandit and reinforcement learning settings. At its core is a novel algorithm, (de)centralized Alternating Gradient Descent and Minimization (AltGDmin), designed to address the challenges of non-convex, under-sampled, and constrained problems. This approach enables fast and federated representation learning while effectively tackling data heterogeneity, scalability, communication efficiency, and safety concerns. The project technical plan includes three research thrusts together with several validation activities and an integrated education plan. The first goal is to create a provable few-shot personalized federated multi-task representation learning framework for bandit and reinforcement learning. The second goal is to develop a fully decentralized, federated multi-task representation learning framework for bandit and reinforcement learning over a networked architecture among agents. Our approach will eliminate the central server, enhancing scalability, communication efficiency, and robustness by removing single points of failure. The third goal is to develop an innovative safe multi-task representation learning framework to learn optimal policies while incorporating the safety constraints. The main application domains of interest are control and automation in smart farms, which will guide the problem formulation and validate the algorithms using real-world implementation, testing, and numerical experiments. The overarching goal of the integrated education plan is to provide a pathway for K-12 to college students to receive rigorous math training and hands-on experience, including coding skills for ML-based intelligent system design. 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

Algorithms are the building blocks of modern computation that underlies much of the technology on which our society depends. In particular, the successes of modern machine learning systems rely on our ability to solve large-scale computational tasks via efficient algorithms, such as for solving optimization problems for training neural networks, and drawing random samples from probability distributions. Algorithms in practice necessarily operate via discrete iterations, by executing one operation after another. Many algorithms are inspired by the mathematics of dynamical systems, where time flows continuously, but all such derivations have been done manually, often with sophisticated and difficult specialized analysis. A general theory of how to derive algorithms from dynamics that can preserve the desired properties is missing. This project aims to close this gap by developing a rigorous way to systematically translate continuous-time dynamics into discrete-time algorithms that can automatically preserve convergence guarantees, focusing on algorithms for fundamental problems in optimization and sampling motivated by machine learning applications. The outcome of this project will benefit many fields in which computational tasks such as optimization and sampling are used, by providing researchers and practitioners a higher-level language involving dynamical systems to design and reason about algorithms. This project also provides training opportunities for students, outreach events to popularize the dynamical perspective of algorithm design, and a collaborative effort to compile a catalog of dynamics and algorithms. This project will strengthen the matching parallel structures between dynamics and algorithms for optimization and sampling, by studying two complementary directions. In the first direction, this project studies sampling problems as optimization problems on the space of probability distributions. Many recent results in sampling have been derived from this perspective, in particular for the basic greedy methods analogous to gradient descent. This project aims to complete the translation from optimization to sampling by deriving algorithmic results for various sampling problems as inspired by the optimization results, including for constrained sampling, accelerated sampling, and sampling with stronger mixing time guarantees. This project will combine ideas and tools from dynamical systems, geometry, probability, and information theory to extract concrete algorithmic principles to design new algorithms in practice. In the second direction, this project aims to formalize the robust parallel structures between dynamics and algorithms with matching convergence guarantees. This project will study the mechanics of translating dynamics to algorithms, and investigate why certain ideas are successful in the translation, including the idea of speeding up time to achieve faster convergence rates, the optimality of accelerated methods for optimization and sampling, and the geometric structure of the space of algorithms and dynamics that allows such robust parallel structures to be exhibited. 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

Today, brain plasticity is understood to include not only changes in how nerve cells communicate with each other, but also with cells responsible for producing the protective covering around nerve fibers, a substance known as myelin. When animals are isolated for long periods, it can affect regions of the brain that are involved in social behaviors. One of these changes is a reduction in myelin, and while this decrease is linked to social difficulties, it is not clear how it happens at a neuro-mechanistic level. The researcher of this project use a combination of novel gene-editing techniques, brain mapping tools, and behavioral tests to determine whether and how a brain nerve chemical called dopamine, which plays a role in a wide variety of behaviors, is involved in the reduction of myelin by social isolation. The findings will not only help advance the field’s understanding of the effects of social isolation but also of how dopamine influences brain changes seen in disorders like schizophrenia and addiction, which are linked to problems with myelin. This project also directly benefits local New York City communities by providing access to laboratory training and encouraging baccalaureate degree completion among a wide variety of students at a local college. Students’ families will be included in STEM outreach, increasing the likelihood of support and encouragement for the students’ academic and research endeavors, as well as a potential impact on younger relatives who may then be more likely to view STEM as a viable future direction. Social isolation is highly relevant across the animal kingdom due to the crucial role of social connections for all species. Across species, chronic social isolation causes experience-dependent myelination – specifically hypomyelination – in the medial prefrontal cortex (mPFC), a brain region critical for mammalian social behaviors. Evidence shows that this effect is associated with social deficits; however, the neurochemical mechanisms mediating this myelin plasticity are not known. Experiments in this proposal test the hypothesis that mesocortical dopamine mediates the effects of chronic adult social isolation on mPFC hypomyelination in mice. The investigators combine genetic and viral strategies (Cre-dependent adeno-associated virus vectors and CRISPR-Cas9 mutagenesis) with neuroanatomical approaches and behavioral experiments to test the predictions that: (1) anatomical and functional interactions occur between mesocortical dopamine terminals and local oligodendrocyte lineage cells; (2) midbrain dopamine neuron activity is necessary and/or sufficient to regulate mPFC myelination; and (3) midbrain dopamine neurons modulate mPFC myelination during chronic social isolation of adult animals. Data in support of the hypothesis that mesocortical dopamine mediates the effects of chronic adult social isolation on mPFC hypomyelination holds implications beyond social isolation because of dopamine’s wide-ranging roles in behavior. By identifying a role for dopamine in mPFC myelin plasticity, this research may guide the development of new pharmacotherapies that target myelin for the treatment of dopamine-related neuropsychiatric diseases, such as schizophrenia and addiction, which are known to be associated with altered myelin 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

Wearable medical imaging devices capable of continuously monitoring health in daily life will significantly advance healthcare. Traditional medical imaging equipment, although powerful, tends to be bulky, expensive, and energy-intensive, limiting its accessibility and usability. This innovative research proposes to miniaturize a photoacoustic imaging (PAI) system—a promising technology combining ultrasound and light—to visualize deep tissue functions. By developing a novel signal encoding and decoding framework, the hundreds of detection channels needed in conventional PAI can be drastically reduced to just one. This advance makes it possible to create lightweight, affordable, and energy-efficient wearable imaging devices. Such devices will greatly benefit telemedicine, enhance early diagnosis and timely intervention of diseases, particularly in underserved and remote communities, and ultimately contribute to better patient outcomes. Additionally, the project incorporates educational activities designed to inspire undergraduate and high-school students to pursue careers in medical device innovation and STEM fields. The technical goal of this CAREER project is to develop a novel spatiotemporal signal encoding and decoding framework enabling the miniaturization of PAI. Conventional PAI requires arrays of hundreds to thousands of ultrasonic detectors for high-resolution imaging, resulting in large, expensive, and power-hungry equipment. To overcome these limitations, the project will implement a specialized encoding medium (hardware) to uniquely tag photoacoustic waves with distinct spatiotemporal signatures, enabling serial detection by a single-element detector. Along with this hardware innovation, a mathematical decoding algorithm (software) will be developed to reconstruct the original signals, effectively creating over one thousand virtual detection channels from a single physical detector. This integrated hardware-software approach will maintain imaging performance comparable to traditional multi-detector systems while dramatically reducing the device size, cost, and energy consumption by one to two orders of magnitude. Successful completion of this research will provide foundational knowledge for designing compact and efficient wearable PAI and ultrasound devices, promising wearable applications for clinical diagnostics and healthcare delivery. 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

This Faculty Early Career Development Program (CAREER) grant funds research that enables general purpose robotic systems that can change their forms to optimally achieve a range of functions. This research introduces a systematic framework for synthesizing form and function within a dynamical system, replacing manual motion design with a scalable, tractable, and data-efficient approach. Unlike traditional fixed-form systems, whose shapes are determined at design time and tailored to specific tasks, this research enables next-generation platforms that can continuously morph to select shapes for solving complex multi-stage tasks in an optimal manner, thereby promoting the progress of science, advancing national prosperity and welfare, and securing the national defense. Tightly integrated with the research activities, this grant also funds a comprehensive outreach strategy to engage participants across various educational levels, including K-12 students, schoolteachers, undergraduate students, and graduate students, and to establish a foundation for lasting contributions to robotics theory, system design, and STEM education through layered mentorship and interdisciplinary learning in the United States. Mobile robotic systems involve complex dynamics with high degrees of freedom, hybrid transitions, and sensitivity to contact and the environment. These challenges are magnified in morphable systems, where the configuration space is combinatorially large and time-varying. Overcoming them requires new representations, numerical methods, and control strategies that generalize across shapes and tasks. This research aims to develop a systematic framework for modeling, analyzing, and controlling hybrid dynamical systems with structured morphological variability and to provide theoretical and algorithmic tools that enable scalable co-design of physical form and control across diverse tasks. The research encompasses three thrusts: (1) constructing a unified framework for modeling morphology using symbolic representations of form and symmetry-aware model reduction; (2) characterizing the form-function relationship and constructing a task-based motion library through trajectory optimization, sensitivity-guided continuation, and bifurcation analysis; and (3) developing a novel data-driven hierarchical control strategy that enables rapid adaptation to changes in morphology and tasks, which will be validated on physical robot platforms with diverse morphologies. Beyond robotic systems, this research has potential applications in reconfigurable and automated manufacturing lines, space exploration missions, and senior and medical care centers, where multitasking capability and versatile operation are strongly required. 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

Vulnerability discovery for software security poses significant challenges due to the vast program state space in complex, real-world programs. This project tackles the challenge of vulnerability discovery in software systems through the enhancement of binary symbolic execution, a technique that simplifies the vast and complex landscape of software operations. Despite its potential, symbolic execution requires substantial expert intervention to manage its complexity, making the process cumbersome and prone to errors. By advancing the automation of this technique, the project promises to significantly boost the efficiency and reliability of detecting vulnerabilities. Advances will improve security for critical infrastructure and software systems by reducing susceptibility to cyber-attacks. The results of the research will be integrated into teaching, outreach and capture-the-flag competitions. This project introduces an innovative approach to improve the scalability of binary symbolic execution, a technique essential for detecting software vulnerabilities. The research will develop a system, referred to as SE-bot, which automates the detection process traditionally performed by human cybersecurity experts. This involves analyzing strategies used by experts in handling symbolic execution tools. These strategies will be decomposed into tasks that can be automated using machine learning techniques to predict and address performance bottlenecks. The system will not only replicate current expert strategies but will also proactively prevent issues before they arise. The proposed research includes three components: 1) detecting performance bottleneck using a combination of machine learning and heuristics, 2) mitigating the slowdown through a number of techniques such as path prioritization and partial execution, and 3) predicting performance bottlenecks and mitigating them. 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

This Faculty Early Career Development (CAREER) project supports research that aims to advance robotic tactile sensing technology by enabling robots to actively explore and perceive the physical properties of objects, such as hardness, texture, and slipperiness. This capability is essential for robots to handle complex tasks, ranging from manipulating delicate fabrics to working with irregular objects like sauces or seasoning particles. The research looks to develop innovative frameworks that enable robots to co-optimize their actions and observation models, leading to improved perception and manipulation capabilities. This work has significant societal and educational benefits. Enhanced tactile sensing will expand the range of tasks robots can perform in healthcare, manufacturing, and everyday life, making them more versatile and impactful. Additionally, the project will integrate its findings into new robotics courses, fostering the education of a new generation of roboticists. Openly accessible educational materials will promote broader community engagement and diversity in robotic tactile sensing research, contributing to national prosperity and workforce development. This CAREER project addresses fundamental challenges in robotic tactile sensing by developing methods for active perception that integrate exploratory actions with tactile signal interpretation. Co-optimization frameworks will be formulated to minimize perception uncertainty by refining both exploratory actions and observation models. These frameworks will draw upon insights from psychological studies, physical modeling, and data-driven optimization, enabling robots to perceive complex object properties. The research introduces implicit feature representations for properties that are difficult to measure directly, utilizing these representations for both perception and manipulation. This project will provide the first general solution for using active touch to perceive physical properties and for integrating active touch perception with precise manipulation. By framing the perception-manipulation system as a co-optimization problem, the methods will advance both theoretical understanding and practical applications. A sensorized dexterous hand will validate the approach through challenging tasks involving both solid and formless objects, pushing the boundaries of current robotic perception and manipulation technologies. 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

Lakes provide important services to communities across the United States, including recreation, drinking water, and habitat for fish and wildlife. Yet many lakes are becoming saltier, mainly due to road runoff and sea level rise. While too much salt can be toxic, very low levels of salt can also stress aquatic life. In places like coastal Maine, lakes span a wide range of salinity due to natural factors and human influence. This project studies how changing salt levels affect lake food webs, focusing on Daphnia–a tiny freshwater grazer that helps keep algal blooms in check and serves as food for fish. Daphnia can evolve quickly—within just a few generations—making them a powerful model for studying how rapid adaptation to environmental change can ripple through ecosystems. Understanding how Daphnia cope with salt stress could help explain and predict changes in water clarity and fish populations. In addition to advancing scientific knowledge, this research will create hands-on research opportunities for undergraduate students, integrate local fieldwork and lab experiments into college courses, and partner with state agencies and citizen scientists to support lake stewardship. The project investigates how ecological and evolutionary responses to salinity shape freshwater lake ecosystems. The research integrates field surveys of zooplankton and phytoplankton communities across a natural salinity gradient in coastal Maine lakes, laboratory common garden experiments quantifying local adaptation in Daphnia ambigua populations, and population genomic analyses to characterize spatial and temporal patterns of salt tolerance. Complementary mesocosm experiments manipulate salinity and resource availability to test how salt-driven evolutionary changes in Daphnia influence consumer-resource dynamics, potentially intensifying herbivory with consequent evolutionary feedbacks. By linking phenotypic and genetic variation in a keystone herbivore to community- and ecosystem-level outcomes, this project will contribute to a mechanistic understanding of eco-evolutionary dynamics in nature. Findings will inform both theory in evolutionary ecology and the management of freshwater systems experiencing salinization due to land-use change and rising sea levels. 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

This project introduces HoloStream, a novel system that enables the efficient analysis of large data streams on different types of computing platforms. The project’s novelties include (i) a customizable design that can be tailored to various deployment settings, (ii) mechanisms to tune system configuration according to workload changes, and (iii) techniques to manage resources automatically. The project's broader significance and importance are the potential to make data streaming tools more accessible to everyday users, enabling new applications in areas like smart cities, healthcare, personalized recommendations, and tracking disease outbreaks. The project involves three sets of tasks that address challenges in adaptive resource management and efficient reconfiguration of streaming applications on dataflow systems. First, the investigator designs an adaptive distributed runtime system that can be tailored to the diverse workload characteristics of streaming applications and achieve practical performance on heterogeneous deployments. Second, she introduces online adaptation mechanisms to achieve consistent and correct reconfiguration of streaming applications without downtime. Third, the investigator develops a heterogeneity-aware optimization framework to enable self-management of streaming applications. Project results have the potential to lower the deployment costs of streaming technology for users and providers alike, and inform future research on self-management policies for long-running applications, beyond streaming analytics. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF

Advances in additive manufacturing now enable the fabrication of thin walls and complex geometries with multiple materials. The synthesis and processing of shape memory alloy materials using laser-based additive manufacturing has the potential to create adaptable and responsive complex structures, such as foldable metal origami shapes, which can be used in applications with high strength and temperature requirements with intricate shapes beyond the conventional manufacturing capabilities. This Faculty Early Career Development (CAREER) project will support research that promotes the progress of the science of creating functionally graded shape memory alloy structures by additive manufacturing techniques and studying foldable shape memory origami patterns. Expanding understanding of manufacturing adaptive metal origami structures that can result from this project look to help shape future actuation and control needs in mobility, aerospace, biomedical devices, robotics, construction, and defense industrial sectors. This project has three research thrusts: 1) Modeling and characterizing the microstructure, mechanical, and thermomechanical properties of shape memory alloy material manufactured by laser powder bed fusion with functionally graded material designs along fold lines, (2) Defining the effect of origami features and functionally graded materials during additive manufacturing on origami shape memory alloy pattern folding performance, and (3) Investigating computational topological models for origami shape memory alloy structures by including material properties and manufacturing constraints within an origami pattern enumerating algorithm. The integration of materials and manufacturing constraints in the unfolding analysis of origami shape memory alloy looks to enable the optimization of adaptive origami shape memory alloy geometries to precisely change shape upon stimulus. The project introduces and boosts interest in STEM research among undergraduate students by developing and integrating educational activities focused on developing Course-based Undergraduate Research Experiences (Additive Manufacturing-CURES) in the manufacturing and materials courses. The educational plan includes holding science workshops and outreach activities through the Toledo Excel Institute for K-8 students, in addition to two planned workshops at the Imagination Station Children’s Museum at Toledo. 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.