Search NASA⌕ Search

SEARCH · Search NASA

Results for “Software Quality Models”

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.

572 records · Page 32
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Multiple social platforms reveal actionable signals for software vulnerability awareness: A study of GitHub, Twitter and Reddit

Software vulnerabilities are flaws in computer systems that leave users open to attack. In many cases, these vulnerabilities go unnoticed and remain unresolved in codebases. Thus, awareness of software vulnerabilities among the public is crucial to ensure effective cybersecurity practices, the development of high quality software, and ultimately national security. This awareness can be better understood by studying the spread and evolution of software vulnerability discussions in online communities. This work is the first to evaluate and contrast how discussions about software vulnerabilities spread on three social platforms -- Twitter, GitHub, and Reddit. To lay the groundwork, we showcase a novel fundamental framework for measuring information spread that identifies the spread mechanisms and observables across platforms, the units of information, and the groups of measurements that can be applied to focus on a specific phenomena e.g., information cascades. We then analyze and contrast social network topologies for three example social networks and measure the scale and speed of the spread of discussion of specific vulnerabilities to understand how far and how widely they spread, how many users participate in discussions, and the duration of their spread. To demonstrate the awareness of more impactful software vulnerabilities, a subset of our analysis focuses on vulnerabilities targeted during recent major cyber attacks as well as vulnerabilities exploited by advanced persistent threat groups. We discover that usually, vulnerability discussions start on GitHub, before occurring on Twitter and Reddit. While studying how some user-level and content-level characteristics influence vulnerability spread, we observe that Twitter discussions started by users predicted to be humans have larger size, breadth, depth, adoption rate, lifetime, and structural virality compared to those started by users predicted to be bots. On Reddit, we contrast the differences in thread structure that originate from posts with positive, negative and neutral polarity. We find that posts that are positive have larger, deeper and wider discussions compared to negative and neutral posts. We anticipate the results of our analysis to not only increase the understanding of software vulnerability awareness but also inform models for simulating information spread across multiple social environments online.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1993

Statistical analysis of trigonometric parallaxes

The present project was initiated with two specific medium-term goals: first, to develop a novel approach, based on global modeling and maximum likelihood, to the study of databases of stellar data, with specific attention to the results to be obtained by the Hipparcos mission; and second, to apply and test the methodology developed to existing ground-based data. The long-term goal was of course to integrate the methodology and the ground-based data into a global study of the results of the Hipparcos mission, which were expected to be made available in 1995 or 1996. The closing of the project is due to the recent change of home institution of the PI, who has moved from the University of Illinois to Johns Hopkins University. As a consequence of this move, the three scientists involved in this project will be in close proximity, thereby hopefully improving their ability to collaborate and increasing the productivity of the project. Continued funding for the project has been requested from Johns Hopkins University. This report describes briefly the results obtained so far both on the technical aspect of software development and on the scientific side of applications to existing ground-based data. Significant progress has been made on both counts, with several papers published in (or submitted to) refereed journals and in conference proceedings. Because it is hoped that the project can be successfully continued with NASA support, the report of the progress in each area includes also an assessment of how the current results fit in the expected continuation of the project. Our results to date include: code development (essentially completed); a detailed study of the kinematics and dynamics of stars escaping from the Hyades cluster, relevant to the question of membership in the Hyades; a study of the kinematics and luminosity calibration of nearby dwarfs; an assessment of the quality of the photometry included in the Hipparcos Input Catalog; and two studies of properties of nearby clusters, including a moving-cluster determination of the distance to Praesepe. The bibliography includes three papers submitted to refereed journals, two of which have already been published, and four contributions to conference proceedings. Finally, the work so far has also provided a very good introduction to stellar dynamics and astrometry for an undergraduate student, with educational benefits that had not been foreseen in the original proposal.

Casertano, Stefano↗
NASA NTRS2005

Failure Assessment

Three questions to which software developers want accurate, precise answers are "How can the software system fail?", "mat bad things will happen if the software fails?t', and "How many failures will the software experience?". Numerous techniques have been devised to answer these questions; three of the best known are: 1) Software Fault Tree Analysis (SFTA) 2) Software Failure Modes, Effects, and Criticality Analysis (SFMECA 3) Software Fault/Failure Modeling. SFTA and SFMECA have been successfully used to analyze the flight software for a number of robotic planetary exploration missions, including Galileo, Cassini, and Deep Space 1. Given the increasing interest in reusing software components from mission to mission, one of us has developed techniques for reusing the corresponding portions of the SFTA and SFMECA, reducing the effort required to conduct these analyses. SFTA has also been shown to be effective in analyzing the security aspects of software systems; intrusion mechanisms and effects can easily be modeled using these techniques. The Bi- Directional Safety Analysis (BDSA) method combines a forward search (similar to SFMECA) from potential failure modes to their effects, with a backward search (similar to SFTA) from feasible hazards to the contributing causes of each hazard. BDSA offers an efficient way to identify latent failures. Recent work has extended BDSA to product-line applications such as flight-instrumentation displays and developed tool support for the reuse of the failure-analysis artifacts within a product line. BDSA has also been streamlined to support those projects having tight cost and/or schedule constraints for their failure analysis efforts. We discuss lessons learned from practice, describe available tools, and identi@ some future directions for the topic. A substantial amount of research has been devoted to estimating the number of failures that a software system will experience during test and operations, as well as the number of faults that have been inserted into that system during its development. One of us has found that the amount of structural change to a system during its development is strongly related to the number of faults inserted into it. Using techniques requiring no additional effort on the part of the development organization, the required measurements of structural evolution can be easily obtained from a development effort's configuration management system and readily transformed into an estimate of fault content. So far, structure-fault relationships have been identified for source code; current work seeks to examine artifacts available earlier in the lifecycle to determine if similar relationships between structure and fault content can be found. In particular, relationships between requirements change requests and the number of faults inserted into the implemented system would provide a significant improvement in our ability to control software quality during the early development phases.

fault tree↗
NASA NTRS2001

Issues in deep space radiation protection

The exposures in deep space are largely from the Galactic Cosmic Rays (GCR) for which there is as yet little biological experience. Mounting evidence indicates that conventional linear energy transfer (LET) defined protection quantities (quality factors) may not be appropriate for GCR ions. The available biological data indicates that aluminum alloy structures may generate inherently unhealthy internal spacecraft environments in the thickness range for space applications. Methods for optimization of spacecraft shielding and the associated role of materials selection are discussed. One material which may prove to be an important radiation protection material is hydrogenated carbon nanofibers. c 2001. Elsevier Science Ltd. All rights reserved.

short duration↗
DOE OSTI2021

Large-scale Gravitational Lens Modeling with Bayesian Neural Networks for Accurate and Precise Inference of the Hubble Constant

We investigate the use of approximate Bayesian neural networks (BNNs) in modeling hundreds of time delay gravitational lenses for Hubble constant (H 0 ) determination. Here, our BNN was trained on synthetic Hubble Space Telescope quality images of strongly lensed active galactic nuclei with lens galaxy light included. The BNN can accurately characterize the posterior probability density functions (PDFs) of model parameters governing the elliptical power-law mass profile in an external shear field. We then propagate the BNN-inferred posterior PDFs into an ensemble H 0 inference, using simulated time delay measurements from a plausible dedicated monitoring campaign. Assuming well-measured time delays and a reasonable set of priors on the environment of the lens, we achieve a median precision of 9.3% per lens in the inferred H 0 . A simple combination of a set of 200 test lenses results in a precision of 0.5 km s –1 Mpc –1 (0.7%), with no detectable bias in this H 0 recovery test. The computation time for the entire pipeline—including the generation of the training set, BNN training and H0 inference—translates to 9 minutes per lens on average for 200 lenses and converges to 6 minutes per lens as the sample size is increased. Being fully automated and efficient, our pipeline is a promising tool for exploring ensemble-level systematics in lens modeling for H 0 inference.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

Data and scripts from: “Denoising autoencoder for reconstructing sensor observation data and predicting evapotranspiration: noisy and missing values repair and uncertainty quantification”

This data package includes data and scripts from the manuscript “Denoising autoencoder for reconstructing sensor observation data and predicting evapotranspiration: noisy and missing values repair and uncertainty quantification”.The study addressed common challenges faced in environmental sensing and modeling, including uncertain input data, missing sensor observations, and high-dimensional datasets with interrelated but redundant variables. Point-scaled meteorological and soil sensor observations were perturbed with noises and missing values, and denoising autoencoder (DAE) neural networks were developed to reconstruct the perturbed data and further predict evapotranspiration. This study concluded that (1) the reconstruction quality of each variable depends on its cross-correlation and alignment to the underlying data structure, (2) uncertainties from the models were overall stronger than those from the data corruption, and (3) there was a tradeoff between reducing bias and reducing variance when evaluating the uncertainty of the machine learning models.This package includes:(1) Four ipython scripts (.ipynb): “DAE_train.ipynb” trains and evaluates DAE neural networks, “DAE_predict.ipynb” makes predictions from the trained DAE models, “ET_train.ipynb” trains and evaluates ET prediction neural networks, and “ET_predict.ipynb” makes predictions from trained ET models.(2) One python file (.py): “methods.py” includes all user-defined functions and python codes used in the ipython scripts.(3) A “sub_models” folder that includes five trained DAE neural networks (in pytorch format, .pt), which could be used to ingest input data before being fed to the downstream ET models in ‘ET_train.ipynb” or ‘ET_predict.ipynb’.(4) Two data files (.csv). Daily meteorological, vegetation, and soil data is in “df_data.csv”, where “df_meta.csv” contains the location and time information of “df_data.csv”. Each row (index) in “df_meta.csv” corresponds to each row in “df_data.csv”. These data files are formatted to follow the data structure requirements and be directly used in the ipython scripts, and they have been shuffled chronologically to train machine learning models. The meteorological and soil data was collected using point sensors between 2019-2023 at(4.a) Three shrub-dominated field sites in East River, Colorado (named “ph1”, “ph2” and “sg5” in “df_meta.csv”, where “ph1” and “ph2” were located at PumpHouse Hillslopes, and “sg5” was at Snodgrass Mountain meadow) and(4.b) One outdoor, mesoscale, and herbaceous-dominated experiment in Berkeley, California (named “tb” in “df_meta.csv”, short for Smartsoils Testbed at Lawrence Berkeley National Lab).- See "df_data_dd.csv" and "df_meta_dd.csv" for variable descriptions and the Methods section for additional data processing steps. See "flmd.csv" and "README.txt" for brief file descriptions.- All ipython scripts and python files are written in and require PYTHON language software.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Quantum annealing algorithms for Boolean tensor networks

Abstract Quantum annealers manufactured by D-Wave Systems, Inc., are computational devices capable of finding high-quality heuristic solutions of NP-hard problems. In this contribution, we explore the potential and effectiveness of such quantum annealers for computing Boolean tensor networks. Tensors offer a natural way to model high-dimensional data commonplace in many scientific fields, and representing a binary tensor as a Boolean tensor network is the task of expressing a tensor containing categorical (i.e., $$\{0, 1\}$$ { 0 , 1 } ) values as a product of low dimensional binary tensors. A Boolean tensor network is computed by Boolean tensor decomposition, and it is usually not exact. The aim of such decomposition is to minimize the given distance measure between the high-dimensional input tensor and the product of lower-dimensional (usually three-dimensional) tensors and matrices representing the tensor network. In this paper, we introduce and analyze three general algorithms for Boolean tensor networks: Tucker, Tensor Train, and Hierarchical Tucker networks. The computation of a Boolean tensor network is reduced to a sequence of Boolean matrix factorizations, which we show can be expressed as a quadratic unconstrained binary optimization problem suitable for solving on a quantum annealer. By using a novel method we introduce called parallel quantum annealing, we demonstrate that Boolean tensor’s with up to millions of elements can be decomposed efficiently using a DWave 2000Q quantum annealer.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Volumetric Rendering on Wavelet-Based Adaptive Grid

Numerical modeling of physical phenomena frequently involves processes across a wide range of spatial and temporal scales. In the last two decades, the advancements in wavelet-based numerical methodologies to solve partial differential equations, combined with the unique properties of wavelet analysis to resolve localized structures of the solution on dynamically adaptive computational meshes, make it feasible to perform large-scale numerical simulations of a variety of physical systems on a dynamically adaptive computational mesh that changes both in space and time. Volumetric visualization of the solution is an essential part of scientific computing, yet the existing volumetric visualization techniques do not take full advantage of multi-resolution wavelet analysis and are not fully tailored for visualization of a compressed solution on the wavelet-based adaptive computational mesh. Our objective is to explore the alternatives for the visualization of time-dependent data on space-time varying adaptive mesh using volume rendering while capitalizing on the available sparse data representation. Two alternative formulations are explored. The first one is based on volumetric ray casting of multi-scale datasets in wavelet space. Rather than working with the wavelets at the finest possible resolution, a partial inverse wavelet transform is performed as a preprocessing step to obtain scaling functions on a uniform grid at a user-prescribed resolution. As a result, a solution in physical space is represented by a superposition of scaling functions on a coarse regular grid and wavelets on an adaptive mesh. An efficient and accurate ray casting algorithm is based just on these coarse scaling functions. Additional details are added during the ray tracing by taking an appropriate number of wavelets into account based on support overlap with the interpolation point, wavelet coefficient magnitude, and other characteristics, such as opacity accumulation (front to back ordering) and deviation from frontal viewing direction. The second approach is based on complementing of wavelet-based adaptive mesh to the traditional Adaptive Mesh Refinement (AMR) mesh. Both algorithms are illustrated and compared to the existing volume visualization software for Rayleigh-Benard thermal convection and electron density data sets in terms of rendering time and visual quality for different data compression of both wavelet-based and AMR adaptive meshes.

Vezolainen, Alexei V.↗
NASA NTRS1994

Electronic photography at NASA Langley Research Center

The field of photography began a metamorphosis several years ago which promises to fundamentally change how images are captured, transmitted, and output. At this time the metamorphosis is still in the early stages, but already new processes, hardware, and software are allowing many individuals and organizations to explore the entry of imaging into the information revolution. Exploration at this time is prerequisite to leading expertise in the future, and a number of branches at LaRC have ventured into electronic and digital imaging. Their progress until recently has been limited by two factors: the lack of an integrated approach and the lack of an electronic photographic capability. The purpose of the research conducted was to address these two items. In some respects, the lack of electronic photographs has prevented application of an integrated imaging approach. Since everything could not be electronic, the tendency was to work with hard copy. Over the summer, the Photographics Section has set up an Electronic Photography Laboratory. This laboratory now has the capability to scan film images, process the images, and output the images in a variety of forms. Future plans also include electronic capture capability. The current forms of image processing available include sharpening, noise reduction, dust removal, tone correction, color balancing, image editing, cropping, electronic separations, and halftoning. Output choices include customer specified electronic file formats which can be output on magnetic or optical disks or over the network, 4400 line photographic quality prints and transparencies to 8.5 by 11 inches, and 8000 line film negatives and transparencies to 4 by 5 inches. The problem of integrated imaging involves a number of branches at LaRC including Visual Imaging, Research Printing and Publishing, Data Visualization and Animation, Advanced Computing, and various research groups. These units must work together to develop common approaches to image processing and archiving. The ultimate goal is to be able to search for images using an on-line database and image catalog. These images could then be retrieved over the network as needed, along with information on the acquisition and processing prior to storage. For this goal to be realized, a number of standard processing protocols must be developed to allow the classification of images into categories. Standard series of processing algorithms can then be applied to each category (although many of these may be adaptive between images). Since the archived image files would be standardized, it should also be possible to develop standard output processing protocols for a number of output devices. If LaRC continues the research effort begun this summer, it may be one of the first organizations to develop an integrated approach to imaging. As such, it could serve as a model for other organizations in government and the private sector.

Holm, Jack M.↗
NASA NTRS2017

UAV Inspection of Electrical Transmission Infrastructure with Path Conformance Autonomy and Lidar-Based Geofences NASA Report on UTM Reference Mission Flights at Southern Company Flights November 2016

Flights at low altitudes in close proximity to electrical transmission infrastructure present serious navigational challenges: GPS and radio communication quality is variable and yet tight position control is needed to measure defects while avoiding collisions with ground structures. To advance unmanned aerial vehicle (UAV) navigation technology while accomplishing a task with economic and societal benefit, a high voltage electrical infrastructure inspection reference mission was designed. An integrated air-ground platform was developed for this mission and tested in two days of experimental flights to determine whether navigational augmentation was needed to successfully conduct a controlled inspection experiment. The airborne component of the platform was a multirotor UAV built from commercial off-the-shelf hardware and software, and the ground component was a commercial laptop running open source software. A compact ultraviolet sensor mounted on the UAV can locate 'hot spots' (potential failure points in the electric grid), so long as the UAV flight path adequately samples the airspace near the power grid structures. To improve navigation, the platform was supplemented with two navigation technologies: lidar-to-polyhedron preflight processing for obstacle demarcation and inspection distance planning, and trajectory management software to enforce inspection standoff distance. Both navigation technologies were essential to obtaining useful results from the hot spot sensor in this obstacle-rich, low-altitude airspace. Because the electrical grid extends into crowded airspaces, the UAV position was tracked with NASA unmanned aerial system traffic management (UTM) technology. The following results were obtained: (1) Inspection of high-voltage electrical transmission infrastructure to locate 'hot spots' of ultraviolet emission requires navigation methods that are not broadly available and are not needed at higher altitude flights above ground structures. (2) The sensing capability of a novel airborne UV detector was verified with a standard ground-based instrument. Flights with this sensor showed that UAV measurement operations and recording methods are viable. With improved sensor range, UAVs equipped with compact UV sensors could serve as the detection elements in a self-diagnosing power grid. (3) Simplification of rich lidar maps to polyhedral obstacle maps reduces data volume by orders of magnitude, so that computation with the resultant maps in real time is possible. This enables real-time obstacle avoidance autonomy. Stable navigation may be feasible in the GPS-deprived environment near transmission lines by a UAV that senses ground structures and compares them to these simplified maps. (4) A new, formally verified path conformance software system that runs onboard a UAV was demonstrated in flight for the first time. It successfully maneuvered the aircraft after a sudden lateral perturbation that models a gust of wind, and processed lidar-derived polyhedral obstacle maps in real time. (5) Tracking of the UAV in the national airspace using the NASA UTM technology was a key safety component of this reference mission, since the flights were conducted beneath the landing approach to a heavily used runway. Comparison to autopilot tracking showed that UTM tracking accurately records the UAV position throughout the flight path.

Moore, Andrew J.↗
DOE OSTI2021

Dynamic Building Load Control to Facilitate High Penetration of Solar Photovoltaic Generation (Final Technical Report)

Solar photovoltaic (PV) resources are the most common form of distributed generation in residential and commercial customer premises within electric distribution networks. A higher penetration of PV generation in distribution circuits will impose challenges on maintaining service voltages within the range of industry standards, power quality, and power flow. Buildings consume 74% of the electricity produced in the United States, and a significant portion of the building load is dispatchable, making them responsive to electrical grid needs. Oak Ridge National Laboratory—in collaboration with Southern Company; the University of Tennessee, Knoxville; and the Georgia Institute of Technology—is examining the PV integration issues in distribution-level electrical grids and developing integrated demand-side control and communication systems to enable responsive loads. The proposed responsive loads mechanism performs renewable generation following to increase the penetration of solar PV within each feeder. The specific objectives of this project are to (1) examine distribution-level PV integration scenarios to understand requirements, (2) undertake an end-to-end simulation-based design of a distributed control strategy of loads geographically near the PV generation asset to minimize the effect on the distribution feeder, (3) deploy and demonstrate the control technology developed in partnership with utilities, and (4) perform a scalability analysis at the utility scale. This 3-year integrated project aims to develop, demonstrate, and validate demand-side control technology to enable increased the penetration of renewables while mitigating challenges that arise due to their intermittency. Activities in Budget Period (BP) 1 focused on a literature review and the formal design of a control system for integrating local distribution with generation and loads. The team used modeling and simulation to evaluate the impact of varying buildings loads, variable PV generation, and power flow dynamics on the distribution circuit. The dynamic models developed in BP 1 were used in BP 2 to develop a model-based control design and a test bed. The test bed has enabled the simulation-based testing and comparison of different control designs and formulations applied to different configurations of the distribution grid, PVs, and building loads. The control approaches developed in BP 2 were implemented in BP 3 in the form of hardware deployed at the Central Baptist Church (CBC) in Knoxville, Tennessee, for testing and evaluation. The outcome of this project was the development and demonstration of open-source, low-cost, low-touch sensing and control retrofits to distributed PV generation and building loads that, in a coordinated fashion, provide the load-shaping response needed to integrate high levels of renewable penetration. This research addresses the target metrics by dynamically controlling a load with solar generation variability to minimize the extent of two-way power flow, enhance reliability, facilitate high PV penetration (>100% of peak load in a line segment), and generate scalable software and hardware solutions adaptable to any penetration levels. The research and development activities are focused and designed to be impactful within the relevant 2020 targets time frame.An accurate open-source integration simulation framework for end-to-end control design was developed and deployed at the CBC facility for testing and evaluation. This final report provides a detailed review of the technical results achieved during this 3-year integrated project. A novel spectral analysis of PV data is demonstrated to derive the requirements of the control design. A detailed simulation-based analysis of PV integration at increasing penetration levels is presented using 1 year of PV data to demonstrate the impact on the distribution circuits. Two different control strategies were developed and demonstrated via simulation to track variable PV generation with adaptive load dispatch. The report concludes with a summary of accomplishments and recommendations for a path forward.

14 SOLAR ENERGY↗
DOE OSTI2021

Data from: "Warming and provenance limit tree recruitment across and beyond the elevation range of subalpine forest"

This data package contains data used to support conclusions drawn in “Warming and provenance limit tree recruitment across and beyond the elevation range of subalpine forest”, by Kueppers et al. 2017. Data were collected in field sites within the Alpine Treeline Warming Experiment (ATWE), located on Niwot Ridge, on the eastern slope of the Colorado Rocky Mountains, USA. Files containing geospatial data are also included, to provide additional locational context.There are four document formats associated with this archive: three comma-separated values (.csv) files, three Microsoft Excel (.xlsx) files, one .pdf data user’s guide, four keyhole markup language (.kml) files, and a compressed folder containing seven ESRI shapefiles (.shp). The .csv files can be opened using any simple text-editor software, R, or Microsoft Excel. The .xlsx files can only be opened using Microsoft Excel. The .kml file can be opened by Google Earth and Google Maps, and the shapefiles can be opened with any GIS application compatible with the file type, such as ESRI’s ArcGIS, and QGIS.We provide two versions of the seedling data file: “PIEN_PIFLseedlings20150522_20150525rev12222020.csv/.xlsx” (hereafter PIEN_PIFLseedlings2015) and “PIEN_PIFLseedlings20160408rev12222020.csv/.xlsx” (hereafter PIEN_PIFLseedlings2016). PIEN_PIFLseedlings2015 contains the data we used in the paper. PIEN_PIFLseedlings2016 contains an updated version of these data that includes sampling from later years. The main differences between the two files lie in the columns titled “k[YEAR],” which describe the number of seedlings that were killed in a particular year. In PIEN_PIFLseedlings2016, there also is an additional year of data for k2015, and k2014 also has additional data input for the 2014 cohort. Additionally, in years 2010-2014, there are minor differences in the number of seedlings killed -- in as few as 0 plots (in 2011) to as many as 5 plots (in 2014) -- due to errors in data input that were rectified in later years.------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------Upslope range shifts by subalpine tree species are a widely anticipated effect of climate change. Climate niche models predict subalpine forests to expand upslope, given more suitable growing conditions for adult trees. However, these models do not take into account climates required for successful seedling recruitment and establishment, an essential element for expansion. Further, localized upper treeline populations are hypothesized to contain favorable traits for colonizing the alpine. To test these expectations and to expand our knowledge of seedling recruitment under climate change, we designed a common garden, climate-warming experiment spread across an elevation gradient at Niwot Ridge in the Colorado Rocky Mountains. We focus on two widespread Western North American species, Engelmann spruce (Picea engelmannii Parry ex. Engelm) and limber pine (Pinus flexilis James), which occur at treeline. While the former is considered a late-seral species more tolerant of shade, limber pine is a shade-intolerant pioneer species able to establish on infertile sites.Every autumn, seeds of the two species were collected from high- (3370 m–3570 m) and low-provenance (2910–3240 m) sources close to the experimental sites and sown in our plots. A subset of plots were heated and another subset watered over the summer months to offset the effects of warming. Across five years, we found that seeds originating from low elevation recruited more strongly for both species, although this provenance difference diminished by the fourth year for Engelmann spruce, likely due to small sample sizes. Despite the recruitment of low-provenance seed, warming treatments decreased recruitment at all elevations. Combining this with the likeliness and availability of lower-quality, high provenance seed moving upslope at the treeline, tree migration into the alpine may be slowed. Overall, our findings suggest that the hardier limber pine is likely to become a more significant species in subalpine forest communities in the future, while the more sensitive Engelmann spruce may experience range contraction.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning

Large language models (LLMs) are increasingly adapted to downstream tasks via reinforcement learning (RL) methods like Group Relative Policy Optimization (GRPO), which often require thousands of rollouts to learn new tasks. We argue that the interpretable nature of language often provides a much richer learning medium for LLMs, compared to policy gradients derived from sparse, scalar rewards. To test this, we introduce GEPA (Genetic-Pareto), a prompt optimizer that thoroughly incorporates natural language reflection to learn high-level rules from trial and error. Given any AI system containing one or more LLM prompts, GEPA samples trajectories (e.g., reasoning, tool calls, and tool outputs) and reflects on them in natural language to diagnose problems, propose and test prompt updates, and combine complementary lessons from the Pareto frontier of its own attempts. As a result of GEPA's design, it can often turn even just a few rollouts into a large quality gain. Across six tasks, GEPA outperforms GRPO by 6% on average and by up to 20%, while using up to 35x fewer rollouts. GEPA also outperforms the leading prompt optimizer, MIPROv2, by over 10% (e.g., +12% accuracy on AIME-2025), and demonstrates promising results as an inference-time search strategy for code optimization. We release our code at https://github.com/gepa-ai/gepa.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Assessment of the Griffin Reactor Multiphysics Application Using the Empire Micro Reactor Design Concept

In late 2019, INL and ANL agreed to jointly develop the reactor physics code named Griffin based on the integration of the two code suites, MAMMOTH/Rattlesnake (INL) and MC2 - 3/PROTEUS (ANL). Griffin is being developed based on the MOOSE framework and MOOSE quality assurance procedures. This decision was made to be able to allow DOE-NE to efficiently invest funding to this area and to provide effective and timely support for existing and potential users; the latter includes industry and government organizations who are developing various types of advanced reactors in the near and long term. Since MAMMOTH/Rattlesnake has been developed based on the MOOSE framework, the INL/ANL Griffin development team agreed to build Griffin beginning with a merger of MAMMOTH and Rattlesnake into a single code and moving forward by implementing capabilities from the PROTEUS suite into Griffin. Moving forward, both ANL and INL efforts are equally invested in the Griffin project, with management support, to provide an advanced reactor multiphysics tool to assist in reactor design, optimization, and safety analysis. Much work remains in moving Griffin forward to migrate PROTEUS capabilities and to optimize performance to meet user needs. The main objective of this work is to assess the current status of Griffin capabilities in terms of performance and accuracy, to determine priorities for PROTEUS migration, and to identify capabilities and features to improve for supporting the code integration effort. For this assessment, the Empire micro reactor problem that was developed in the ARPA-E MEITNER program was selected as an advance reactor concept of interest to the technical community. The Empire reactor problem was expanded from its original incomplete specification to be a small heat-pipe-cooled micro reactor core with ~113 cm radius and 70 cm in height, composed of 18 fuel assemblies, 12 control drums, and beryllium radial and axial reflectors. In the current model, using 5 cm axial reflectors specified in the original Empire assembly model, more than 10% of neutrons leak axially and through the empty center safety hole, as well as through heat pipe channels in fuel assembly elements that extend through the top reflector region. Several calculation models of the core were defined for systematic assessment, including 2-D and 3-D fuel assemblies and whole cores with cylindrical boundaries. Cross sections were generated using Serpent 2, and meshes were produced using the Argonne mesh tool or the INL neutronics meshing tools combined with CUBIT. Cross sections and meshes were converted to the ISOXML and Exodus formats, respectively, so that Griffin and PROTEUS could use consistent data for solving the reactor problems. With the prepared cross sections and meshes, PROTEUS was run first to ensure that all input data were correctly generated and input options in terms of angle, mesh, and energy group were accurately determined. Comparisons against Serpent 2 solutions were made in terms of eigenvalue and pin power. The same calculations and comparisons were then conducted using Griffin. For the fuel assembly and whole core problems, the PROTEUS eigenvalues agreed well with reference Serpent 2 solutions within 100 and 30 pcm, respectively, and pin power differences relative to Serpent 2 were overall less than 2.2% and RMS 0.8% for the whole core models. This indicated that all input data were properly prepared. Using the same data, Griffin was run selecting the SAAF-CFEM SN solver with Legendre-Gaussian quadrature and NDA and DSA for acceleration. It was found that the SAAF-CFEM solver of Griffin required finer meshes to achieve eigenvalue and pin power solutions in good agreement with Serpent 2, consequently requiring more memory requirement and longer computation time. On the other hand, the SPH-Diffusion 2-D core calculations performed using Griffin were able to recover the exact eigenvalue from the reference Serpent 2 solutions, resulting in a pin-power distribution with an RMS of 0.6% and maximum absolute difference of less than 1.4%. The runtimes for SPH-Diffusion for the 2-D core were less than 3 minutes on 40 cores. During this evolution of this evaluation, many updates were made in Griffin by the Griffin development team of INL (focusing on software updates) and ANL (reviewing and supporting software updates) to complete this assessment. Observations from the code assessment are presented in the conclusion section of this report, followed by a discussion of recommendations for future work.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗