Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Domain knowledge”

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 415 records · Page 23
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Physics-informed KNN milling stability model with process damping effects

This paper describes a k-nearest neighbors, or KNN, model for milling stability including process damping effects. A physics-based, frequency domain milling stability solution is used to generate the training data, but does not incorporate process damping effects. The data set is then updated using limited tests to capture the process damping behavior. A “stair step” approach is used to select the test points, where a first spindle speed-axial depth combination is selected based on the physics-based stability map, subsequent tests are defined using the previous test result, and data points are updated by knowledge of process damping behavior and the test results. Furthermore, the KNN modeling approach demonstrates the ability to predict both stable and unstable results, including process damping behavior.

42 ENGINEERING↗
DOE OSTI2025

DeFault: DEep‐Learning‐Based FAULT Delineation Using the IBDP Passive Seismic Data at the Decatur CO2 Storage Site

Abstract The carbon capture, utilization, and storage (CCUS) framework is an essential component in reducing greenhouse gas emissions, with its success hinging on the comprehensive knowledge of subsurface geology and geomechanics. Passive seismic event relocation and fault detection offer vital insights into subsurface structures and the ability to monitor fluid migration pathways. Accurate identification and localization of seismic events, however, face significant challenges, including the necessity for high‐quality seismic data and advanced computational methods. To address these challenges, we introduce a novel deep learning method, , specifically designed for passive seismic source relocation and fault delineating for passive seismic monitoring projects. By leveraging data domain‐adaptation, allows us to train a neural network with labeled synthetic data and apply it directly to field data. Using , the passive seismic sources are automatically clustered based on their recording time and spatial locations, and subsequently, faults and fractures are delineated accordingly. We demonstrate the efficacy of on a field case study involving injection related microseismic data from Decatur, Illinois area. Our approach accurately and efficiently relocated passive seismic events, identified faults and could aid in potential damage induced by seismicity. Our results highlight the potential of as a valuable tool for passive seismic monitoring, emphasizing its role in ensuring CCUS project safety. This research bolsters the understanding of subsurface characterization in CCUS, illustrating machine learning’s capacity to refine these methods. Ultimately, our work has significant implications for CCUS technology deployment, an essential strategy in combating climate change. Plain Language Summary In our quest to tackle climate change, we use a strategy known as carbon capture, utilization, and storage (CCUS) to keep greenhouse gases out of the atmosphere. This strategy relies heavily on our ability to understand what's happening deep under the earth's surface. To make sure we store super critical safely, we need to accurately map out the geological structure, especially faults, but this is tough without high‐quality data and complex computer programs. We've developed a new tool called “DeFault,” which uses advanced machine learning to improve how we find and map these underground features. “DeFault” is smart enough to learn from numerically simulated data and then apply what it’s learned to real‐world situations. It groups together seismic activity—tiny tremors and shifts in the earth—based on when and where they happen, which helps us spot where there might be cracks or faults. We tested “DeFault” in Illinois, where CO 2 is injected underground, and it successfully pinpointed where these tremors occurred and mapped out the faults, helping to prevent accidents accurately in the future. Our study shows that “DeFault” will be a powerful ally in making CCUS safer and more effective, especially for the Illinois Basin Decatur Project. Key Points Faults and fractures introduced by carbon storage can be monitored by passive seismicity DeFault algorithm enables an automatic process for accurate and efficient passive seismic event locating and clustering

58 GEOSCIENCES↗
DOE OSTI2020

Skeletides: A Modular, Simplified Physical Model of Protein Secondary Structure

Three-dimensional (3D) models are essential for visualization and conceptual understanding of complex architectures such as protein structure. Although there is a plethora of software platforms that allow digital depictions of protein structure at an atomic level in silico, physical models are needed to convey an intuitive understanding of biomolecular architecture. However, it is a challenge to represent all the relevant features of proteins in a single physical model due to their sheer structural complexity. Here, we describe a modular protein model that focuses only on representation of the secondary structure—the underlying structural skeleton. The simplified model consists of amino acid units, which can be linked together to reproduce the two most fundamental structural features of protein secondary structure: the relative positions of the amino acid alpha carbon atoms, and the intra-main chain hydrogen bonding pattern. We use 3D printing and magnets to create a set of three modular amino acid building blocks, which when linked together into a chain, can faithfully represent alpha helices, beta sheets, and turns. These simple to make models can be used to quickly assemble the alpha carbon trace of an entire protein domain, which conveys a tactile experience of the complexity of protein skeletal architecture. These models highlight the modularity of protein structure: where a single structural unit, the amino acid, can be linked together to form a larger, regular secondary structure. These models also have a propensity to spontaneously organize into alpha helices and beta sheets, as demonstrated by their ability to autonomously assemble when placed in a circulating water tank. These models provide a missing educational tool to expand knowledge of protein structure, foster deeper insight into protein folding, and inspire greater interest in biomacromolecular architecture.

36 MATERIALS SCIENCE↗
DOE OSTI2026

BrickQA: Bridging the Semantic Gap in Building Operations with Dynamic Graph Exploration

While standardized ontologies like the Brick schema address data heterogeneity in Building Automation Systems (BAS), accessing this semantic data remains a challenge as domain experts often lack the expertise to formulate complex SPARQL queries. To bridge this gap, we present BrickQA, a Large Language Model (LLM)-based framework that translates natural language into executable SPARQL queries through structured query decomposition, dynamic schema exploration, and inline validation. BrickQA utilizes an iterative reasoning agent to actively navigate graph topology through dynamic exploration actions without requiring exhaustive context injection or model fine-tuning. This approach effectively mitigates hallucinations, particularly in large-scale building knowledge graphs. Empirical evaluation on BuildingQA, a standardized benchmark, demonstrates that BrickQA significantly outperforms ReAct baselines, delivering a 0.291–0.355 absolute F1 improvement while achieving 3 × –12.7 × higher token cost-efficiency. Beyond these metrics, the framework maintains structural fidelity across heterogeneous buildings and remains resilient to ambiguous queries without requiring site-specific fine-tuning. Furthermore, a case study on operational analytics validates the framework’s capability to handle temporal and aggregation constraints, effectively transforming abstract semantic models into actionable facility management insights.1

Ko, Yun-Dam↗
DOE OSTI2023

Rapid Computational Identification of Therapeutic Targets for Pathogens

Biological threats continue to persist and evolve as an important challenge to national security. There are multiple ways in which novel viral pathogens could emerge to pose a serious threat to human health. This project developed a pathogen target identification tool that can rapidly respond to a novel or emerging viral biological threat. A set of computational tools were developed that provide detailed information on the newly sequenced genes, their protein products and the drug target sites for the proteins that are best suited for biological countermeasure development. Three key innovations were developed in the project. 1) Development of a new extensive database of protein pocket structures with structure-based search algorithms to rapidly link novel protein targets with the complete collection of previously experimentally solved protein structures. 2) A novel clustering pipeline was introduced to group matching structures and associated small-molecule binding ligands into a consensus protein pocket with the associated small-molecule chemotypes predicted to fit in the pocket site. The matching experimentally solved structures were used to inform the value of different target sites. 3) Where there are viral protein targets with pockets structurally matched to similar human proteins, a biological knowledge graph, which links molecular interactions with human disease, was used to further assess the potential negative impact of a viral protein target with similarities to human proteins that could have important off target side effects. In total, the project produced a new resource for rapid and detailed assessment of promising targets for countermeasures, reflecting the ongoing wet lab, clinical, and computational data being collected. These capabilities will improve the ability to respond to a biological threat in multiple domains.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

In-situ Characterizations of Microstructural Degradation of Perovskite Solar Cells

Rapid progress has been achieved in perovskite solar cells (PSCs), and their efficiencies have improved from 3.8 % to 24.2 % in less than a decade. With low-cost processing, PSCs have shown exciting photovoltaic (PV) properties, such as effective optical absorption, a long carrier lifetime, and unique defect tolerance. While recent studies demonstrated improved stability up to 100 days, PSC technology is still challenged to meet the stringent industry requirements for commercialization. Despite considerable efforts, the underlying physical mechanisms for the inferior stability of PSCs are not well understood. One reason for this divergence is that many established measurement techniques (e.g., quantum efficiency, photoluminescence) probe the properties on length scales far greater than that of electronic and/or structural inhomogeneity (i.e., < 1 μm near grain boundaries) and therefore characterize convoluted and/or averaged properties. Ion/electron beam-based techniques have been extensively used to access the microstructures of PSCs, enabling atomic/nanoscale structural, chemical, optical, and electrical characterizations. For example, focused ion beam (FIB) milling produces an atomically smooth surface that minimizes the artifacts attributed to the surface roughness. FIB techniques can also create a well-defined cross-section of PSCs without mechanical damage in a physical cleaving sample preparation. While powerful, there are some concerns about possible beam damage of inorganic-organic perovskites via chemical-bond breakage and local heating. This project aims to comprehensively understand how the microstructural/interfacial properties of PSCs (e.g., Methylammonium Lead Iodide [MAPbI 3 ]) are modified under the irradiating ion beams. Specifically, we investigate the sub-surface properties of PSCs before and after Ar-ion beam injections. Kelvin probe force microscopy (KPFM) measures the contact potential differences (CPDs). Photoluminescence (PL) microscopy in conjunction with Finite-Difference Time-Domain (FDTD) simulations infers the formation of a “dead layer” (< 15 nm) on the subsurface of MAPbI 3 during Ar+ milling processes while preserving the initial bulk properties. The x-ray photoemission spectroscopy (XPS) confirms this modified surface is a lead-rich and iodine-deficient surface. We initiate customizing in-situ measurement setup while measuring the local optical and electrical properties of PSC under thermal (cooling, heating) and light stressors. Our results provide in-depth knowledge of the ion-beam impact on metal-halide perovskites and how this modified sub-surface impacts their properties under accelerated stressors of light and heat. Intensive Monte Carlo simulations of an electron beam interacting with PSCs provide the beam energy distribution in PSCs, proposing possible measurement conditions of using e-beam with minimizing beam damage. Our in-situ measurement platform can accommodate the diverse architecture of PSC devices for studying deterioration mechanisms under mixed environmental stressors.

14 SOLAR ENERGY↗
DOE OSTI2024

MetaPoL: Immersive VR based Indoor Patterns of Life (PoL) and Anomalies Data Generation for Insider Threat Modeling in Nuclear Security

Insider threats are perhaps the most serious challenges that nuclear and radiological security systems face. Insiders pose such a great threat due to their access, authority, and knowledge, granting them opportunities to bypass dedicated nuclear and radiological security elements. For example, in one of the latest major insider threat incidents to nuclear security, the Doel-4 nuclear powerplant in Belgium suffered a shutdown, the threat of nuclear materials diversion, and long-term loss of tens of millions of dollars. Seven years of investigation concluded that it was an inside job and attempted sabotage. In this regard, there is an immediate need for R&D and technology integration in the domain of modeling indoor Patterns-of-Life (PoL) and anomaly detection. This can be achieved by using datasets of facility users’ mobility and activity, which can support the design of algorithms for insider threat modeling and detection. However, due to classification, privacy, sensitivity, and safety protocols, such datasets from real physical nuclear reactor facilities are not only hard to share, but also not always feasible to deploy and collect. Aiming to find an alternate solution, our proposed demonstration work - MetaPoL, is the first-ever (for the application space) immersive VR (virtual reality) environment of a real-world secure facility and allows users to move-and-stay through the designed indoor physical layout and also encounter NPCs (non-player characters) that emulate other facility users. In the MetaPoL an interactive user performs realistic spatio-temporal movement, dwelling and activities using a Meta Quest Pro VR headset, and that generates high-frequency (in time) high-resolution (in space) indoor spatial-temporal datasets that are valuable for PoL modeling and anomaly detection research specifically for insider threat modeling and detection mission. Such generated realistic, rich in context, and mission specific datasets can boost AI/Machine Learning based research for modeling and detecting insider threats in nuclear security and nonproliferation.

Gunaratne, Chathika↗
DOE OSTI2021

Explaining and predicting human behavior and social dynamics in simulated virtual worlds: reproducibility, generalizability, and robustness of causal discovery methods

Ground Truth program was designed to evaluate social science modeling approaches using simulation test beds with ground truth intentionally and systematically embedded to understand and model complex Human Domain systems and their dynamics Lazer et al. (Science 369:1060–1062, 2020). Our multidisciplinary team of data scientists, statisticians, experts in Artificial Intelligence (AI) and visual analytics had a unique role on the program to investigate accuracy, reproducibility, generalizability, and robustness of the state-of-the-art (SOTA) causal structure learning approaches applied to fully observed and sampled simulated data across virtual worlds. In addition, we analyzed the feasibility of using machine learning models to predict future social behavior with and without causal knowledge explicitly embedded. In this paper, we first present our causal modeling approach to discover the causal structure of four virtual worlds produced by the simulation teams—Urban Life, Financial Governance, Disaster and Geopolitical Conflict. Our approach adapts the state-of-the-art causal discovery (including ensemble models), machine learning, data analytics, and visualization techniques to allow a human-machine team to reverse-engineer the true causal relations from sampled and fully observed data. We next present our reproducibility analysis of two research methods team’s performance using a range of causal discovery models applied to both sampled and fully observed data, and analyze their effectiveness and limitations. We further investigate the generalizability and robustness to sampling of the SOTA causal discovery approaches on additional simulated datasets with known ground truth. Our results reveal the limitations of existing causal modeling approaches when applied to large-scale, noisy, high-dimensional data with unobserved variables and unknown relationships between them. We show that the SOTA causal models explored in our experiments are not designed to take advantage from vasts amounts of data and have difficulty recovering ground truth when latent confounders are present; they do not generalize well across simulation scenarios and are not robust to sampling; they are vulnerable to data and modeling assumptions, and therefore, the results are hard to reproduce. Finally, when we outline lessons learned and provide recommendations to improve models for causal discovery and prediction of human social behavior from observational data, we highlight the importance of learning data to knowledge representations or transformations to improve causal discovery and describe the benefit of causal feature selection for predictive and prescriptive modeling.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Estimating building occupancy: a machine learning system for day, night, and episodic events

Building occupancy research increasingly emphasizes understanding the social and physical dynamics of how people occupy space. Opportunities in the open source domain including social media, Volunteered Geographic Information, crowdsourcing, and sensor data have proliferated, resulting in the exploration of building occupancy dynamics at varying spatiotemporal scales. At Oak Ridge National Laboratory, research into building occupancies through the development of a global learning framework that accommodates exploitation of open source authoritative sources, including governmental census and surveys, journal articles, real estate databases, and more, to report national and subnational building occupancies across the world continues through the Population Density Tables (PDT) project. This probabilistic learning system accommodates expert knowledge, experience, and open-source data to capture local, socioeconomic, and cultural information about human activity. It does so through a systematic process of data harmonization techniques in the development of observation models for over 50 building types to dynamically update baseline estimates and report probabilistic diurnal and episodic building occupancy estimates. This discussion will explore how PDT is implemented at scale and expanded based on the development of observation model classes and will explain how to interpret and spatially apply the reported probability occupancy estimates and uncertainty.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2024

A Privacy-Preserving Cyber Threat Intelligence Sharing System

Cyber Threat Intelligence (CTI) is a key resource for developing defensive strategies against potential cyber adversaries. Entities typically access CTI through open-source platforms, national agencies, or specialized commercial services. However, the bi-directional exchange of CTI is hindered by organizational trust boundaries, which complicate the sharing processes between entities and CTI providers. Centralized CTI services benefit from receiving suspicious cyber observables such as IP addresses, domain names, and email addresses from various entities. The aggregation allows for the correlation of widespread adversarial activities to enhance the alert and response mechanisms across the network of involved parties. Despite these benefits, openly sharing such observables incurs potential legal, regulatory, and reputational risks for the disclosing entities.This paper introduces a system designed to facilitate the secure exchange of cyber observables across trust boundaries without compromising the anonymity of the sharing entities. Here, we propose an architecture that leverages common web protocols alongside zero-knowledge proofs to authenticate members while maintaining anonymity. Additionally, we outline a privacy model tailored for STIX (Structured Threat Information eXpression) cyber observables to minimize the risk of inadvertently disclosing private information. Through our threat models, we assess the privacy implications of our proposed system and demonstrate its potential to enhance collaborative cyber defense efforts without exposing entities to undue risk.

BBS+ Signatures↗
DOE OSTI2020

LAN: A Materials Notation for Two-Dimensional Layered Assemblies

Two-dimensional (2D) layered materials offer intriguing possibilities for novel physics and applications. Before any attempt at exploring the materials space in a systematic fashion, or combining insights from theory, computation, and experiment, a formal description of information about an assembly of arbitrary composition is required. Here, we introduce a domain-generic notation that is used to describe the space of 2D layered materials from monolayers to twisted assemblies of arbitrary composition, existent or not yet fabricated. The notation corresponds to a theoretical materials concept of stepwise assembly of layered structures using a sequence of rotation, vertical stacking, and other operations on individual 2D layers. Its scope is demonstrated with several example structures using common single-layer materials as building blocks. This work overall aims to contribute to the systematic codification, capture, and transfer of materials knowledge in the area of 2D layered materials.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Structural basis of a shared antibody response to SARS-CoV-2

Molecular understanding of neutralizing antibody responses to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) could accelerate vaccine design and drug discovery. We analyzed 294 anti–SARS-CoV-2 antibodies and found that immunoglobulin G heavy-chain variable region 3-53 (IGHV3-53) is the most frequently used IGHV gene for targeting the receptor-binding domain (RBD) of the spike protein. Co-crystal structures of two IGHV3-53–neutralizing antibodies with RBD, with or without Fab CR3022, at 2.33- to 3.20-angstrom resolution revealed that the germline-encoded residues dominate recognition of the angiotensin I converting enzyme 2 (ACE2)–binding site. This binding mode limits the IGHV3-53 antibodies to short complementarity-determining region H3 loops but accommodates light-chain diversity. These IGHV3-53 antibodies show minimal affinity maturation and high potency, which is promising for vaccine design. Knowledge of these structural motifs and binding mode should facilitate the design of antigens that elicit this type of neutralizing response.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Hypergraph Analytics of Domain Name System Relationships

We report on the use of novel mathematical methods in hypergraph analytice over a large quantity of DNS data. Hypergraphs generalize graphs, as used in network science, to better model complex multiway relations in cyber data. Specifically, casting DNS data from Georgia Tech's ActiveDNS repository as hypergraphs allows us to fully represent the interactions between {\em collections} of domains and IP addresses. To facilitate large-scale analytics, we fielded an analytical pipeline of two capabilities. HyperNetX (HNX) is a Python package for the exploration and visualization of hypergraphs, acting as a frontend. For the backend, the Chapel HyperGraph Library (CHGL) is a library for high performance hypergraph analytics written in the exascale programming language Chapel. CHGL was used to process gigascale DNS data, performing compute-intensive calculations for data reduction and segmentation. Identified portions are then sent to HNX for both exploratory analysis and knowledge discovery targeting known tactics, techniques, and procedures.

Joslyn, Cliff A.↗
DOE OSTI2021

ARENA: Asynchronous Reconfigurable Accelerator Ring to Enable Data-Centric Parallel Computing

The next generation HPC and data centers are likely to be reconfigurable and data-centric due to the trend of hardware specialization and the emergence of data-driven applications. In this work, we propose ARENA – an asynchronous reconfigurable accelerator ring architecture as a potential scenario on how the future HPC and data centers will be like. Despite using the coarse-grained reconfigurable arrays (CGRAs) as the substrate platform, our key contribution is not only the CGRA-cluster design itself, but also the ensemble of a new architecture and programming model that enables asynchronous tasking across a cluster of reconfigurable nodes, so as to bring specialized computation to the data rather than the reverse. We presume distributed data storage without asserting any prior knowledge on the data distribution. Hardware specialization occurs at runtime when a task finds the majority of data it requires are available at the present node. In other words, we dynamically generate specialized CGRA accelerators where the data reside. The asynchronous tasking for bringing computation to data is achieved by circulating the task token, which describes the dataflow graphs to be executed for a task, among the CGRA cluster connected by a fast ring network. Evaluations on a set of HPC and data-driven applications across different domains show that ARENA can provide better parallel scalability with reduced data movement (53.9 percent). Compared with contemporary compute-centric parallel models, ARENA can bring on average 4.37× speedup. The synthesized CGRAs and their task-dispatchers only occupy 2.93mm 2 chip area under 45nm process technology and can run at 800MHz with on average 759.8mW power consumption. ARENA also supports the concurrent execution of multi-applications, offering ideal architectural support for future high-performance parallel computing and data analytics systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Origins of heat transport anisotropy in MoTe2 and other bulk van der Waals materials

Knowledge of how heat flows anisotropically in van der Waals (vdW) materials is crucial for thermal management of emerging 2D materials devices and design of novel anisotropic thermoelectric materials. Despite the importance, anisotropic heat transport in vdW materials is yet to be systematically studied and is often presumably attributed to anisotropic speeds of sound in vdW materials due to soft interlayer bonding relative to covalent in-plane networks of atoms. In this work, we investigate the origins of the anisotropic heat transport in vdW materials, through time-domain thermoreflectance (TDTR) measurements and first-principles calculations of anisotropic thermal conductivity of three different phases of MoTe 2 . MoTe 2 is ideal for the study due to its weak anisotropy in the speeds of sound. We find that even when the speeds of sound are roughly isotropic, the measured thermal conductivity of MoTe 2 along the c-axis is 5–8 times lower than that along the in-plane axes. We derive meaningful characteristic heat capacity, phonon group velocity, and relaxation times from our first principles calculations for selected vdW materials (MoTe 2 , BP, h-BN, and MoS 2 ), to assess the contributions of these factors to the anisotropic heat transport. Interestingly, we find that the main contributor to the heat transport anisotropy in vdW materials is anisotropy in heat capacity of the dominant heat-carrying phonon modes in different directions, which originates from anisotropic optical phonon dispersion and disparity in the frequency of heat-carrying phonons in different directions. Further, the discrepancy in frequency of the heat-carrying phonons also leads to ~2 times larger average relaxation times in the cross-plane direction, and partially explains the apparent dependence of the anisotropic heat transport on the anisotropic speeds of sound. This work provides insight into understanding of the anisotropic heat transport in vdW materials.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Call to Action for Global Access to and Harmonization of Quality Information of Individual Earth Science Datasets

Knowledge about the quality of data and metadata is important to support informed decisions on the (re)use of individual datasets and is an essential part of the ecosystem that supports open science. Quality assessments reflect the reliability and usability of data. They need to be consistently curated, fully traceable, and adequately documented, as these are crucial for sound decision- and policy-making efforts that rely on data. Quality assessments also need to be consistently represented and readily integrated across systems and tools to allow for improved sharing of information on quality at the dataset level for individual quality attribute or dimension. Although the need for assessing the quality of data and associated information is well recognized, methodologies for an evaluation framework and presentation of resultant quality information to end users may not have been comprehensively addressed within and across disciplines. Global interdisciplinary domain experts have come together to systematically explore needs, challenges and impacts of consistently curating and representing quality information through the entire lifecycle of a dataset. This paper describes the findings of that effort, argues the importance of sharing dataset quality information, calls for community action to develop practical guidelines, and outlines community recommendations for developing such guidelines. Practical guidelines will allow for global access to and harmonization of quality information at the level of individual Earth science datasets, which in turn will support open science.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2024

Unifying Combinatorial and Graphical Methods in Artificial Intelligence

Recently, a new graph Laplacian, called the inner product Laplacian, was introduced which generalizes many existing Laplacians, including the normalized and combinatorial Laplacian and their weighted variants. The key observation behind the inner product Laplacian is that by defining appropriate inner product spaces on the vertices and edges, the standard Laplacians can be recovered as Hodge Laplacians over the simplicial complex formed by the edges and vertices. These inner product spaces form a natural way to incorporate non-combinatorial information into the definition of a domain-specific Laplacian. In particular, in contrast to current domain-specific weighting schemes which rely solely on edge weights, information regarding the similarity of non-adjacent vertices and arbitrary pairs of edges can be effectively incorporated into the Laplacian. In order to illustrate this approach we consider the problem of calculating the potential energy of an atomistic configuration using Graph Neural Networks. In comparison with start-of-the-art approaches, such as SchNet, our approach replaces a learned (via auto-encoder) representation of the atom types with an inner product space on atoms based on scientific knowledge (e.g., electronegativity). We will illustrate how this approach captures key chemical properties of the molecules and compare the energy calculations with state-of-the-art neural network approaches. However, to compute the resulting Laplacian involves a mixture of sparse and dense matrix computation and yields a dense matrix as the basis for the graph convolution. This dense convolutional kernel necessitates moving away from the standard message passing framework for graph neural networks and increases the computational cost of applying the kernel. In order to mitigate these costs we investigate means of leveraging the mixed sparse and dense computations to reduce the overall computational cost and how these approaches can be automatically transferred to energy efficient hardware (e.g., field programmable gate arrays (FPGAs)).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Yes, No, Maybe So: Human Factors Considerations for Fostering Calibrated Trust in Foundation Models Under Uncertainty

High-stakes analytical environments require analysts to evaluate evidence and generate conclusions to inform critical decisions often under conditions of uncertainty. Probabilistic decision-making based on incomplete or inaccurate information can reduce productivity, compromise national interests, and endanger public safety. Researchers are developing expert systems built on foundation models (FMs) to support analysts’ decision-making processes by enabling human-artificial intelligence (AI) teaming, in part through the quantification and expression of uncertainty information. As FMs continue to mature, it is imperative to correspondingly consider analysts’ needs for appropriately interpreting and using uncertainty information. However, prior research indicates that it remains unclear how analysts engage with FM-generated uncertainty information and the extent to which these interactions influence trust in, and reliance on, expert systems. We plan to review the state of the science and conduct an exploratory, qualitative study to (a) understand how properly communicated uncertainty can foster calibrated trust and appropriate reliance and (b) identify approaches for effectively conveying FM-generated uncertainty information during analytical workflows. We will administer semi-structured interviews with analysts from a specific high-stakes analytical environment to collect their current experiences with job-related uncertainty and their impressions when viewing FM-generated uncertainty information. During the interview protocol, participants will be presented with several different FM outputs and invited to discuss their thoughts and beliefs about the uncertainty information displayed. Participants may provide insights into how trust and reliance may be influenced by uncertainty. The results of this study will help us to better understand how analysts currently interpret and use uncertainty information. Our findings may inform human factors recommendations for effectively conveying uncertainty information to foster calibrated trust in, and appropriate reliance on, expert systems. Interaction designers and FM developers can use this knowledge to enhance human-AI teaming and ensure the responsible deployment of FM-based expert systems in analytical workflows.

97 MATHEMATICS AND COMPUTING↗