Search NASA⌕ Search

SEARCH · Search NASA

Results for “Constrained learning”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 253 records · Page 14
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Machine Learning for Mapping Multipactor Susceptibility in RF Systems: Capabilities and Generalization Constraints

Multipactor is a surface-driven electron avalanche phenomenon that degrades the performance and reliability of radio-frequency (RF) systems in particle accelerator and vacuum electronics applications. Multipactor behavior in a given device structure is conventionally assessed through susceptibility charts, which provide a parameter-space characterization of the instability. In this work, we assess the capabilities of machine-learning (ML) models to learn and predict such susceptibility charts and analyze the constraints governing their generalization across materials. Using a simulation-derived dataset spanning six distinct secondary-electron-yield material profiles in a canonical two-surface planar geometry, we train supervised regression models and artificial neural networks to predict the time-averaged electron growth rate, δavg, across the relevant parameter space. Model performance is evaluated using metrics that explicitly probe the structure of susceptibility charts, including Intersection over Union, Structural Similarity Index, and correlation analysis. Tree-based ensemble models outperform neural-network models in reconstructing susceptibility regions and in generalizing across material domains. Principal-component analysis reveals disjoint material feature distributions, indicating that the piecewise mode structure of multipactor susceptibility is difficult to represent with a single global model and that generalization is constrained by data coverage rather than by model complexity. An exhaustive reduced-coverage study further shows that sparse material-space coverage can yield mean performance in the same general range but producing large variability in the susceptibility-region overlap. These results clarify the capabilities of ML-based surrogate models for parameter-space characterization of multipactor discharge. They also provide guidance for their appropriate use in RF system design.

43 PARTICLE ACCELERATORS↗
DOE OSTI2025

Adversarial Binaries: AI-guided Instrumentation Methods for Malware Detection Evasion

Adversarial binaries are executable files that have been altered without loss of function by an AI agent in order to deceive malware detection systems. Progress in this emergent vein of research has been constrained by the complex and rigid structure of executable files. Although prior work has demonstrated that these binaries deceive a variety of malware classification models which rely on disparate feature sets, a consensus as to the best approach has not been reached, either in terms of the optimization algorithms or the instrumentation methods. Furthermore, although inconsistencies in the data sets, target classifiers, and functionality verification methods make head-to-head comparisons difficult, here we extract lessons learned and make recommendations for future research.

malware obfuscation↗
DOE OSTI2024

Supporting ARPA-E Power Grid Optimization (Final Report)

Pacific Northwest National Laboratory (PNNL), Arizona State University (ASU), Georgia Institute of Technology (Georgia Tech), Los Alamos National Laboratory (LANL), National Renewable Energy Laboratory (NREL), Texas A&M University (TAMU), The University of Texas at Austin (UT), and the University of Wisconsin-Madison (UW-M) supported the ARPA-E Grid Optimization (GO) Competition by providing a common problem formulation, data format, datasets, evaluation mechanism, scoring, rules, and results that resulted in the awarding of $\$9.24$ million dollars to teams from academia, industry, and national labs for solving three sets of increasingly difficult non-linear, security- constrained AC Optimal Powerflow (AC-OPF) optimization problems in order to increase the efficiency of the US Electric Grid. It is estimated that a 1% increase in efficiency can save $\$1$ billion. Current industry practices typically use a linear DC model (DC-OPF) in order solve the OPF problem within the time constraints of the operation schedule. The GO Competition challenges the best power engineers, mathematicians, and computer scientists to make possible operational decisions based on accurate physical models. To accomplish this, the GO Competition created a series of Challenges and funded teams to produce the best solver. Challenge 1 was to solve the security constrained Alternating Current Optimal Power Flow (ACOPF) problem. Challenge 2 extended that to by adding adjustable transformer tap ratios, phase shifting transformers, switchable shunts, price-responsive demand, ramp rate constrained generators and loads, and fast-start unit commitment (UC). Furthermore, Challenge 2 was a maximization problem while Challenge 1 was a minimization problem. While Challenge 3 was being developed, the entrants were invited to find better solutions to the Challenge 2 synthetic datasets with no restrictions on time, hardware, or algorithms. The Challenge 2 solutions turned out to be very good. Challenge 3 expanded the Challenge 2 problem further by using multiperiod dynamic markets, including advisory models for extreme weather events, day-ahead markets, and the real-time markets with an extended look-ahead. These problems included active bid-in demand and topology optimization. Together the Challenges used nearly 30 million CPU hours. Since each team was working on the same problem, using the same data, and running on the same hardware, fair comparisons could be drawn as to the best solver. The datasets were varied enough, however, that the best solver for one dataset was not necessarily the best at another, so cumulative scores were used. The process was managed by the PNNL maintained website https://GOCompetition.energy.gov, where Entrants could find information about the problem, the data, the rules, submit their solver for evaluation, and see the scores of all the competing teams on a Leaderboard. Interest was world-wide but only American teams were eligible for prizes. The Competition has produced 34 journal articles 115 papers and been cited over 500 times in the literature, including 12 dissertations (4 from foreign countries; Columbia (2), Germany, and Italy) and 3 from the DOE ExaScale project. Software developed by Pearl Street Technologies for Challenges 1 and 2 is now deployed by Southwest Power Pool (SPP) and Midcontinent Independent Service Operator (MISO). Other teams have received inquiries from venture capitalists. Google DeepMind has thanked the Competition for making the datasets developed for the Competition public. They are using it to train machine learning models. The larger datasets have billions of unknowns to be solved for, but only a small percent matter in the final solution. Knowing what unknowns are important can dramatically speedup the solution.

24 POWER TRANSMISSION AND DISTRIBUTION↗
NASA NTRS1993

Experiences with thin film filter development for the Extreme Ultraviolet Explorer

The design, development, and optimization of the thin film filters used on the Extreme Ultraviolet Explorer (EUVE) Satellite to define the EUV wavelength bandpasses of the individual instruments was a complicated task. The bandpasses had to be optimized for the astrophysical goals of the EUVE mission and constrained by the strong geocoronal EUV background emission. Materials with optical constants that met these requirements had to be found and tested. In many cases these materials were not compatible or were not strong enough to survive the intense vibrations of a rocket launch. Other effects, such as photoelectron 'halo' produced in the filters, were not discovered until flight qualification. The final set of flight filters included: lexan/boron, aluminum/carbon, titanium/antimony/aluminum, and tin/silicon monoxide. This paper discusses the lessons learned in the development of these filters, including the optimization process, material interactions and problems, calibration techniques, vibration susceptibility, thermal tests, and photoelectron emission. We feel the experiences gained over the last 10 years creating the filter sets for EUVE will be invaluable for future missions that use thin film filters.

Vallerga, J. V.↗
NASA NTRSDate not supplied

Construction of a Fluid Flowfield from Discrete Point Data using Machine Learning

Many verification and validation procedures in aerospace engineering involve the comparison of computational fluid dynamics (CFD) data to experimental results from sources like wind tunnel tests. However, an incongruity exists between the data available from these sources: flow visualization is available by default in computational data, whereas in most experimental setups the available data is far more discrete and far more limited: integrated forces and moments, discrete pressure and temperature probes, etc. When differences exist between quantities of interest like lift and drag coefficients, the lack of full-field flow data from the experiments complicates most attempts to reconcile why the different data sources disagree. To this end, a shallow neural network, constrained by certain fluid flow properties, was trained to approximate flow field snapshots given only discrete data like that available in a wind tunnel test. The constructed snapshots, even for complex incompressible fluid flows, were found to agree at the large scales with the true flow fields. With this tool, researchers can more readily and easily understand why quantities of interest differ between their experimental and computational datasets. This in turn improves the resulting data's uncertainty measures.

Yury Lebedev↗
NASA NTRSDate not supplied

Construction of a Fluid Flowfield from Discrete Point Data using Machine Learning

Many verification and validation procedures in aerospace engineering involve the comparison of computational fluid dynamics (CFD) data to experimental results from sources like wind tunnel tests. However, an incongruity exists between the data available from these sources: flow visualization is available by default in computational data, whereas in most experimental setups the available data is far more discrete and far more limited: integrated forces and moments, discrete pressure and temperature probes, etc. When differences exist between quantities of interest like lift and drag coefficients, the lack of full-field flow data from the experiments complicates most attempts to reconcile why the different data sources disagree. To this end, a shallow neural network, constrained by certain fluid flow properties, was trained to approximate flow field snapshots given only discrete data like that available in a wind tunnel test. The constructed snapshots, even for complex incompressible fluid flows, were found to agree at the large scales with the true flow fields. With this tool, researchers can more readily and easily understand why quantities of interest differ between their experimental and computational datasets. This in turn improves the resulting data's uncertainty measures.

Yury Lebedev↗
DOE OSTI2025

Next generation Arctic vegetation maps: Aboveground plant biomass and woody dominance mapped at 30 m resolution across the tundra biome

The Arctic is warming faster than anywhere else on Earth, placing tundra ecosystems at the forefront of global climate change. Plant biomass is a fundamental ecosystem attribute that is sensitive to changes in climate, closely tied to ecological function, and crucial for constraining ecosystem carbon dynamics. However, the amount, functional composition, and distribution of plant biomass are only coarsely quantified across the Arctic. Therefore, we developed the first moderate resolution (30 m) maps of live aboveground plant biomass (g m −2 ) and woody plant dominance (%) for the Arctic tundra biome, including the mountainous Oro Arctic. We modeled biomass for the year 2020 using a new synthesis dataset of field biomass harvest measurements, Landsat satellite seasonal synthetic composites, ancillary geospatial data, and machine learning models. Additionally, we quantified pixel-wise uncertainty in biomass predictions using Monte Carlo simulations and validated the models using a robust, spatially blocked and nested cross-validation procedure. Observed plant and woody plant biomass values ranged from 0 to ∼6000 g m −2 (mean ≈ 350 g m −2 ), while predicted values ranged from 0 to ∼4000 g m −2 (mean ≈ 275 g m −2 ), resulting in model validation root-mean-squared-error (RMSE) ≈ 400 g m −2 and R 2 ≈ 0.6. Our maps not only capture large-scale patterns of plant biomass and woody plant dominance across the Arctic that are linked to climatic variation (e.g., thawing degree days), but also illustrate how fine-scale patterns are shaped by local surface hydrology, topography, and past disturbance. By providing data on plant biomass across Arctic tundra ecosystems at the highest resolution to date, our maps can significantly advance research and inform decision-making on topics ranging from Arctic vegetation monitoring and wildlife conservation to carbon accounting and land surface modeling.

Climate change↗
DOE OSTI2026

Learned adaptive properties for mitigation of weight perturbations in embedded spiking networks

Recent years have seen an increased importance of neural network inference in edge-based scenarios, which impose size and power constraints requiring novel computing devices. These same edge scenarios may require operating over long periods of time, or exposure to extreme environments, resulting in a drift of neural network weights that cause degraded performance. In searching for ways to develop neural network approaches that perform robustly under these conditions, we propose a biologically-inspired mechanism for the dynamic adaptation of within-neuron parameters that is guided by a global context signal carrying information about perturbations and variability in incoming stimuli. Specifically, we demonstrate that adaptive voltage thresholds or neuronal time constants, when informed by a global context signal, can enable network-level mechanisms to recover from perturbed synaptic weights. Consistent with prior literature, the context-modulated approach is effective for recurrent, but not feedforward networks, by modulating network level dynamics. We demonstrate this approach successfully recovers performance in image classification tasks and spatiotemporal tracking tasks under idealized and Gaussian noise as well as for realistic perturbations from a memristive device when exposed to ionizing radiation. Finally, we discuss how this approach enables the design of robust and energy-efficient neuromorphic systems that perform well, even in resource-constrained scenarios with extreme environments such as edge processing.

context modulation↗
DOE OSTI2025

Randomized Algorithms for Symmetric Nonnegative Matrix Factorization

Symmetric Nonnegative Matrix Factorization (SymNMF) is a technique in data analysis and machine learning that approximates a matrix with a product of a nonnegative, low-rank matrix and it transpose. To design faster and more scalable algorithms for SymNMF we develop two randomized algorithms for its computation. The first method uses randomized matrix sketching to compute an initial low-rank approximation to the input matrix and proceeds to uses this as a low-rank input to rapidly compute a SymNMF. The second methods uses randomized leverage score sampling to approximately solve constrained least squares problems. Many successful methods for SymNMF rely on (approximately) solving sequences of constrained least squares problems. Here, we prove theoretically that leverage score sampling can approximately solve constrained least squares problems to e-accuracy. Finally we demonstrate both methods work in practice by applying them to graph clustering tasks on large real world data sets. These experiments show that our methods approximately maintain solution quality and achieve significant speed ups for both large dense and large sparse problems.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2001

Challenges and Lessons Learned in the Application of Autonomy to Space Operations

NASA's Space Operations Management Office (SOMO) is working toward a goal of providing an integrated infrastructure of mission and data services for space missions undertaken by NASA enterprises. A significant portion of this effort is focused on reducing the cost of these services. We are interested in the potential of autonomy to reduce operations costs. SOMO services support space missions, but are not part of the mission objectives; therefore the level of acceptable risk is very low. In fact, SOMO could be effective ly prevented from applying autonomy if customers merely perceive it as adding risk to their mission(s). We are interested in this workshop from the standpoint of understanding what can be done to realize the potential cost savings due to autonomy while maintaining acceptable risk and serving the needs of our customers. We would like to present our lessons learned so far in adopting autonomy and automation, which we think will contribute to clarifying the challenges facing the use of such technology. SOMO provides services to a diverse and ambitious set of mission customers. Many of these missions are groundbreaking missions for which communications, data, and other operations requirements sometimes cannot be clearly articulated early in the program. This motivates a need for systems that are robust in the face of unanticipated situations so that customer missions are not unreasonably constrained or impacted by "shortcomings" in SOMO services. One of SOMO's primary goals is to realize a paradigm in which SOMO acts as a service provider to organizations that fly space missions for NASA, other government agencies, and even the commercial sector. These organizations purchase SOMO services "by the pound" as customers. We have to provide systems that are not experiments themselves, but rather stable bases from which to do bold experiments. To this end, SOMO also seeks to work closely with industry to see that robust autonomy technology gets infused into products and services for the space industry and beyond. The potential for application of these technologies spans space-based communications networks (e.g. TDRSS) and ground-based assets including communication and tracking antenna systems, data networks, and control centers. There are several problems that are candidates for the application of autonomy, if it can be made reliable enough, including: antenna control, antenna scheduling, communication link scheduling and operation, navigation, attitude determination, fault detection, isolation, and reconfiguration (for spacecraft or ground assets), and mission-level planning and scheduling. Some attempts have been made to apply autonomy and automation in these areas in the past with varying degrees of success. We will present relevant case histories and the lessons inferred from them. Combining this past experience with anticipated future needs, we can clarify the challenges that must be met in order to realize the benefits of autonomy.

Forrest, David J.↗
DOE OSTI2025

Joint state-parameter estimation for the reduced fracture model via the united filter

Here, in this paper, we introduce an effective United Filter method for jointly estimating the solution state and physical parameters in flow and transport problems within fractured porous media. Fluid flow and transport in fractured porous media are critical in subsurface hydrology, geophysics, and reservoir geomechanics. Reduced fracture models, which represent fractures as lower-dimensional interfaces, enable efficient multi-scale simulations. However, reduced fracture models also face accuracy challenges due to modeling errors and uncertainties in physical parameters such as permeability and fracture geometry. To address these challenges, we propose a United Filter method, which integrates the Ensemble Score Filter (EnSF) for state estimation with the Direct Filter for parameter estimation. EnSF, based on a score-based diffusion model framework, produces ensemble representations of the state distribution without deep learning. Meanwhile, the Direct Filter, a recursive Bayesian inference method, estimates parameters directly from state observations. The United Filter combines these methods iteratively: EnSF estimates are used to refine parameter values, which are then fed back to improve state estimation. Numerical experiments demonstrate that the United Filter method surpasses the state-of-the-art Augmented Ensemble Kalman Filter, delivering more accurate state and parameter estimation for reduced fracture models. This framework also provides a robust and efficient solution for PDE-constrained inverse problems with uncertainties and sparse observations.

Bayesian inference↗
DOE OSTI2026

Editorial: Towards the rapid and systematic assessment of vaccine technologies

The COVID-19 pandemic highlighted both the extraordinary potential of modern vaccinology and persistent challenges in how vaccine technologies are assessed. While vaccines can be developed and deployed at unprecedented speed, our ability to predict efficacy in a population is constrained by methodological difficulties, underreporting of negative results, and limited comparability across studies. This editorial introduces a Research Topic that brings together an interdisciplinary collection of experimental, computational, and theoretical contributions spanning multiple pathogens and vaccine platforms. Across these contributions, emerging themes emphasize the need for standardized immunogenicity metrics, transparent reporting including negative findings, and harmonized experimental protocols to support meaningful comparisons. This editorial highlights community practices and shared commitments – supported by researchers, funders, and journals – that could strengthen reproducibility, transparency, and cumulative learning in vaccine research.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

The BTSbot-nearby Discovery of SN 2024jlf: Rapid, Autonomous Follow-up Probes Interaction in an 18.5 Mpc Type IIP Supernova

We present observations of the Type IIP supernova (SN) SN 2024jlf, including spectroscopy beginning just 0.7 days (∼17 hr) after first light. Rapid follow-up was enabled by the new BTSbot-nearby program, which involves autonomously triggering target-of-opportunity requests for new transients in Zwicky Transient Facility data that are coincident with nearby (D < 60 Mpc) galaxies and identified by the BTSbot machine learning model. Early photometry and nondetections shortly prior to first light show that SN 2024jlf initially brightened by >4 mag day −1 , quicker than ∼90% of Type II SNe. Early spectra reveal weak flash ionization features: narrow, short-lived (1.3 < τ[days] < 1.8) emission lines of Hα, He II , and C IV . Assuming a wind velocity of v w = 50 km s −1 , these properties indicate that the red supergiant progenitor exhibited enhanced mass loss in the last year before explosion. We constrain the mass-loss rate to $1{0}^{-4}\lt \dot{M}\,[{M}_{\odot }\,{\mathrm{yr}}^{-1}]\lt 1{0}^{-3}$ by matching observations to model grids from two independent radiative hydrodynamics codes. BTSbot-nearby automation minimizes spectroscopic follow-up latency, enabling the observation of ephemeral early-time phenomena exhibited by transients.

core-collapse supernovae↗
NASA NTRS2002

Simulating "Mars on Earth"

By now, everyone who's heard of the Haughton-Mars Project knows that we travel to Devon Island to learn how people will live and work on Mars. But how do we learn about Mars operations from what happens in the Arctic? We must document our experience--traverses, life in the hab, instrument deployment, communications, and so on. Then we must analyze and formally model what happens. In short, while most scientists are studying the crater, other scientists must be studying the expedition itself. That's what I have done in the past four field seasons. I study field science, both as it naturally occurs at Haughton (unconstrained by a "Mars Sam") and as a constrained experiment using the Flashline Mars Arctic Research Station. During the second week of July 2001, I lived and worked in the hab as part of the Phase 2 crew of six. Besides participating in all activities, I took many photographs and time lapse video. The result of my work will be a computer simulation of how we lived and worked in the hab. It won't be a model of particular people or even my own phase per se, but a pastiche that demonstrates (a proof of concept) that we have appropriate tools for simulating the layout of the hab and daily routines followed by the group and individual scientists. Activities-how people spend their time-are the focus of my observations for building such a simulation model.

Clancey, William J.↗
NASA NTRS2003

Case for Deploying Complex Systems Utilizing Commodity Components

When the International Space Station (ISS) finally reached an operational state, many of the Payload Operations and Integration Facility (POIF) hardware components were reaching end of life, COTS product costs were soaring, and the ISS budget was becoming severely constrained. However, most requirement development was complete. In addition, the ISS program is a fully functioning program with at least fifteen years of operational life remaining. Therefore it is critical that any upgrades, refurbishments, or enhancements be accomplished in realtime with minimal disruptions to service. For these and other reasons, it was necessary to ensure the viability of the POIF. Due to the to the breadth of capability of the POIF (a NASA ground station), it is believed that the lessons to be learned by other complex systems are applicable and any solutions garnered by the POIF are applicable to other complex systems as well. With that in mind, a number of new approaches have been investigated to increase the portability of the POIF and reduce the cost of refurbishment, operations, and maintenance. These new approaches were directed at the Total Cost of Ownership (TCO); not only the refurbishment but also current operational difficulties, licensing, and anticipation of the next refurbishment. Our basic premise is that technology had evolved dramatically since the concept of the POIF ground system and we should leverage our experience on this new technological landscape. Fortunately, Moore's law and market forces have changed the landscape considerably. These changes are manifest in five (5) ways that are particularly relevant to POIF: 1. Complex Instruction Set Computing (CISC) processors have advanced to unprecedented levels of compute capacity with a dramatic cost break, 2. Linux has become a major operating system supported by most vendors on a broad range of platforms, 3. Windows(TradeMark) based desktops are pervasive in the office environment, 4. Stable and affordable WindowsTM development environments and tools are available and offer a rich set of capabilities, 5. The WindowsTM 2000 provides a stable client platform, Therefore, five studies were proposed, developed, and are in the current process of deployment which dramatically reduces the cost of operations, maintenance, refurbishment, and deployment of a ground system. Restating and refining the basic premise stated earlier, it is possible to enhance operations through the replacement of hardware and software components with commodity based items wherever applicable. This will dramatically reduce the overall lifecycle cost of the project. The first study leveraged the POIF S secure, three-tier, web architecture to replace the client workstations with lower cost PC platforms. A second study initiated a review of COTS products to examine the level of added value of each product. This study included replacement of some COTS products with custom code, deletions, substitutions, and consolidation of COTS products. Studies three and four reviewed the server architectures of the data distribution systems and Enhanced HOSC System (EHS) command and telemetry system to propose migration to new platforms, both software and hardware. The final study reviewed current IP communication technologies, developed an operational model for flight operations, and demonstrated that voice over IP was practical and could be integrated into operations.

Bryant, Barry S.↗
DOE OSTI2026

Measurement of muon antineutrino charged current - 0 meson scattering, using the NOvA Near Detector

Antineutrino interaction cross sections are, at present, poorly constrained, particularly regarding the role of multi-nucleon processes such as 2-particle 2-hole (2p2h) interactions. The associated crosssection systematic uncertainties represent a significant challenge for precision oscillation measurements, especially for the next generation of neutrino experiments such as DUNE. We present a new measurement of the muon antineutrino charged-current cross section without mesons in the final state, using the high-statistics data set of the NOvA Near Detector. The analysis employs a cut-based selection enhanced by machine learning techniques to isolate a high-purity sample dominated by quasielastic (QE) and 2p2h interactions. We present the cross section as a function of the kinetic energy and scattering angle of the outgoing muon. We also present measurements of more model-dependent kinematic variables such as the neutrino energy and momentum transfer, to better probe the underlying nuclear physics. The results are compared against various neutrino event generators to test the robustness of current interaction models.

Vockerodt, Kevin John [Ohio State U.; Queen Mary, ↗
DOE OSTI2026

Physics vs structure: A systematic benchmark of learning strategies for multi-zone building thermal dynamics

Recent advances in physics-informed and data-driven machine learning promise improved thermal models for advanced building control, yet there is limited quantitative evidence on when added physics structure and architectural complexity are beneficial. Here, this work presents a systematic benchmark of five representative system identification methods for modeling multi-zone building thermal dynamics: linear state-space models, multi-layer perceptrons, neural state-space models, neural ordinary differential equations, and physically-consistent neural networks. The methods are evaluated across multiple data regimes and zone coupling strategies. Using a high-fidelity multi-zone commercial building emulator, we examine short-term and long-term prediction accuracy, computational efficiency, and ease of development. Our results reveal critical trade-offs between prediction performance, model complexity, and physical consistency. We demonstrate that decoupled, nonlinear black-box models consistently outperform coupled physics-constrained architectures in both predictive accuracy and out-of-distribution robustness in majority of the test cases for the building type considered in the study. Our findings quantify the cost of complexity in building thermal modeling and provide concrete, actionable, scenario-based guidelines for selecting model classes for control-oriented applications.

Building thermal modeling↗
DOE OSTI2024

Dynamically Learning Incentives for Load Control

As electrical generation becomes more distributed and volatile, and loads become more uncertain, controllability of distributed energy resources (DERs), regardless of their ownership status, will be necessary for grid reliability. Grid operators lack direct control over end-users' grid interactions, such as energy usage, but incentives can influence behavior -- for example, an end-user that receives a grid-driven incentive may adjust their consumption or expose relevant control variables in response. A key challenge in studying such incentives is the lack of data about human behavior, which usually motivates strong assumptions, such as distributional assumptions on compliance or rational utility-maximization. In this paper, we propose a general incentive mechanism in the form of a constrained optimization problem -- our approach is distinguished from prior work by modeling human behavior (e.g., reactions to an incentive) as an arbitrary unknown function. We propose feedback-based optimization algorithms to solve this problem that each leverage different amounts of information and/or measurements. We show that each converges to an asymptotically stable incentive with (near)-optimality guarantees given mild assumptions on the problem. Finally, we evaluate our proposed techniques in voltage regulation simulations on standard test beds. We test a variety of settings, including those that break assumptions required for theoretical convergence (e.g., convexity, smoothness) to capture realistic settings. In this evaluation, our proposed algorithms are able to find near-optimal incentives even when the reaction to an incentive is modeled by a theoretically difficult (yet realistic) function.

demand response↗