Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “knowledge graph”

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 217 records · Page 12
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

graphenv: a Python library for reinforcement learning on graph search spaces

Many important and challenging problems in combinatorial optimization (CO) can be expressed as graph search problems, in which graph vertices represent full or partial solutions and edges represent decisions that connect them. Graph structure not only introduces strong relational inductive biases for learning (Battaglia et al., 2018) - in this context, by providing a way to explicitly model the value of transitioning (along edges) between one search state (vertex) and the next - but lends itself to problems both with and without clearly defined algebraic structure. For example, classic CO problems on graphs such as the Traveling Salesman Problem (TSP) can be expressed as either pure graph search or integer programs. Other problems, however, such as molecular optimization, do no have concise algebraic formulations and yet are readily implemented as a graph search (V. et al., 2022; Zhou et al., 2019). Such "model-free" problems constitute a large fraction of modern reinforcement learning (RL) research owing to the fact that it is often much easier to write a forward simulation that expresses all of the state transitions and rewards, than to write down the precise mathematical expression of the full optimization problem. In the case of molecular optimization, for example, one can use domain knowledge alongside existing software libraries to model the effect of adding a single bond or atom to an existing but incomplete molecule, and let the RL algorithm build a model of how good a given decision is by "experiencing" the simulated environment many times through. In contrast, a model-based mathematical formulation that fully expresses all the chemical and physical constraints is intractable. In recent years, RL has emerged as an effective paradigm for optimizing searches over graphs and led to state-of-the-art heuristics for games like Go and chess, as well as for classical CO problems such as the TSP. This combination of graph search and RL, while powerful, requires non-trivial software to execute, especially when combining advanced state representations such as Graph Neural Networks (GNN) with scalable RL algorithms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Track Seeding and Labelling with Embedded-space Graph Neural Networks

To address the unprecedented scale of HL-LHC data, the Exa.TrkX project is investigating a variety of machine learning approaches to particle track reconstruction. The most promising of these solutions, graph neural networks (GNN), process the event as a graph that connects track measurements (detector hits corresponding to nodes) with candidate line segments between the hits (corresponding to edges). Detector information can be associated with nodes and edges, enabling a GNN to propagate the embedded parameters around the graph and predict node-, edge- and graph-level observables. Previously, message-passing GNNs have shown success in predicting doublet likelihood, and we here report updates on the state-of-the-art architectures for this task. In addition, the Exa.TrkX project has investigated innovations in both graph construction, and embedded representations, in an effort to achieve fully learned end-to-end track finding. Hence, we present a suite of extensions to the original model, with encouraging results for hitgraph classification. In addition, we explore increased performance by constructing graphs from learned representations which contain non-linear metric structure, allowing for efficient clustering and neighborhood queries of data points. We demonstrate how this framework fits in with both traditional clustering pipelines, and GNN approaches. The embedded graphs feed into high-accuracy doublet and triplet classifiers, or can be used as an end-to-end track classifier by clustering in an embedded space. A set of post-processing methods improve performance with knowledge of the detector physics. Finally, we present numerical results on the TrackML particle tracking challenge dataset, where our framework shows favorable results in both seeding and track finding.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2022

Multi-objective goal-directed optimization of de novo stable organic radicals for aqueous redox flow batteries

Abstract Advances in the field of goal-directed molecular optimization offer the promise of finding feasible candidates for even the most challenging molecular design applications. One example of a fundamental design challenge is the search for novel stable radical scaffolds for an aqueous redox flow battery that simultaneously satisfy redox requirements at the anode and cathode, as relatively few stable organic radicals are known to exist. To meet this challenge, we develop a new open-source molecular optimization framework based on AlphaZero coupled with a fast, machine-learning-derived surrogate objective trained with nearly 100,000 quantum chemistry simulations. The objective function comprises two graph neural networks: one that predicts adiabatic oxidation and reduction potentials and a second that predicts electron density and local three-dimensional environment, previously shown to be correlated with radical persistence and stability. With no hard-coded knowledge of organic chemistry, the reinforcement learning agent finds molecule candidates that satisfy a precise combination of redox, stability and synthesizability requirements defined at the quantum chemistry level, many of which have reasonable predicted retrosynthetic pathways. The optimized molecules show that alternative stable radical scaffolds may offer a unique profile of stability and redox potentials to enable low-cost symmetric aqueous redox flow batteries.

25 ENERGY STORAGE↗
DOE OSTI2017

Final Report: Sampling-Based Algorithms for Estimating Structure in Big Data.

The purpose of this project was to develop sampling-based algorithms to discover hidden struc- ture in massive data sets. Inferring structure in large data sets is an increasingly common task in many critical national security applications. These data sets come from myriad sources, such as network traffic, sensor data, and data generated by large-scale simulations. They are often so large that traditional data mining techniques are time consuming or even infeasible. To address this problem, we focus on a class of algorithms that do not compute an exact answer, but instead use sampling to compute an approximate answer using fewer resources. The particular class of algorithms that we focus on are streaming algorithms , so called because they are designed to handle high-throughput streams of data. Streaming algorithms have only a small amount of working storage - much less than the size of the full data stream - so they must necessarily use sampling to approximate the correct answer. We present two results: * A streaming algorithm called HyperHeadTail , that estimates the degree distribution of a graph (i.e., the distribution of the number of connections for each node in a network). The degree distribution is a fundamental graph property, but prior work on estimating the degree distribution in a streaming setting was impractical for many real-world application. We improve upon prior work by developing an algorithm that can handle streams with repeated edges, and graph structures that evolve over time. * An algorithm for the task of maintaining a weighted subsample of items in a stream, when the items must be sampled according to their weight, and the weights are dynamically changing. To our knowledge, this is the first such algorithm designed for dynamically evolving weights. We expect it may be useful as a building block for other streaming algorithms on dynamic data sets.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

CheKiPEUQ

Parameter estimation for complex physical problems often suffers from finding 'solutions' that are not physically realistic. The CheKiPEUQ software provides tools for finding physically realistic parameter estimates.and CheKiPEUQ provide tools for making graphs of the parameter positions within parameter space as well as plots of the final simulation results. The primary purpose of the CheKiPEUQ software is to enable more physically realistic parameter estimation from comparing simulations to experiments. Specifically, when prior knowledge is available about the region of parameter space which is physically realistic and when the level of uncertainty from experiment can be estimated. The software is intentionally general and can be used for almost any type of simulation. The software is made in a user friendly manner so that users can simply enter the required data and then run the program, following guidelines provided by the authors along with some trial and error. Examples are provided. Users do not need to understand the methodology that will be namedi in the following sentences. While CheKiPEUQ can be used for conventional parameter estimation and other uses, the primary uses for CheKiPEUQ are: 1) Bayesian Parameter Estimation, 2) Bayesian Model Discrimination, 3) Bayesian Design of Experiments. For more information see the project website, documentation, examples, and related publications.

Savara, Aditya [Oak Ridge National Lab. (ORNL), Oa↗
DOE OSTI2026

ChemGraph as an agentic framework for computational chemistry workflows

Atomistic simulations are essential in chemistry and materials science but remain challenging to run due to the expert knowledge required for the setup, execution, and validation stages of these calculations. We present ChemGraph, an agentic framework powered by artificial intelligence and state-of-the-art simulation tools to streamline and automate computational chemistry and materials science workflows. ChemGraph leverages graph neural network-based foundation models for accurate yet computationally efficient calculations and large language models (LLMs) for natural language understanding, task planning, and scientific reasoning to provide an intuitive and interactive interface. We evaluate ChemGraph across 13 benchmark tasks and demonstrate that smaller LLMs (GPT-4o-mini, Claude-3.5-haiku, Qwen-2.5-14B) perform well on simple workflows, while more complex tasks benefit from using larger models. Importantly, we show that decomposing complex tasks into smaller subtasks through a multi-agent framework enables GPT-4o to reach perfect accuracy and smaller LLMs to match or exceed single-agent GPT-4o's performance in these benchmarks.

Computational chemistry↗
DOE OSTI2024

A Mass‐Conserving‐Perceptron for Machine‐Learning‐Based Modeling of Geoscientific Systems

Although decades of effort have been devoted to building Physical-Conceptual (PC) models for predicting the time-series evolution of geoscientific systems, recent work shows that Machine Learning (ML) based Gated Recurrent Neural Network technology can be used to develop models that are much more accurate. However, the difficulty of extracting physical understanding from ML-based models complicates their utility for enhancing scientific knowledge regarding system structure and function. Here, we propose a physically interpretable Mass-Conserving-Perceptron (MCP) as a way to bridge the gap between PC-based and ML-based modeling approaches. The MCP exploits the inherent isomorphism between the directed graph structures underlying both PC models and GRNNs to explicitly represent the mass-conserving nature of physical processes while enabling the functional nature of such processes to be directly learned (in an interpretable manner) from available data using off-the-shelf ML technology. As a proof of concept, we investigate the functional expressivity (capacity) of the MCP, explore its ability to parsimoniously represent the rainfall-runoff (RR) dynamics of the Leaf River Basin, and demonstrate its utility for scientific hypothesis testing. To conclude, we discuss extensions of the concept to enable ML-based physical-conceptual representation of the coupled nature of mass-energy-information flows through geoscientific systems.

58 GEOSCIENCES↗
DOE OSTI2015

SU-F-BRB-16: A Spreadsheet Based Automatic Trajectory GEnerator (SAGE): An Open Source Tool for Automatic Creation of TrueBeam Developer Mode Robotic Trajectories

Purpose: To automate creation and delivery of robotic linac trajectories with TrueBeam Developer Mode, an open source spreadsheet-based trajectory generation tool has been developed, tested and made freely available. The computing power inherent in a spreadsheet environment plus additional functions programmed into the tool insulate users from the underlying schema tedium and allow easy calculation, parameterization, graphical visualization, validation and finally automatic generation of Developer Mode XML scripts which are directly loadable on a TrueBeam linac. Methods: The robotic control system platform that allows total coordination of potentially all linac moving axes with beam (continuous, step-and-shoot, or combination thereof) becomes available in TrueBeam Developer Mode. Many complex trajectories are either geometric or can be described in analytical form, making the computational power, graphing and programmability available in a spreadsheet environment an easy and ideal vehicle for automatic trajectory generation. The spreadsheet environment allows also for parameterization of trajectories thus enabling the creation of entire families of trajectories using only a few variables. Standard spreadsheet functionality has been extended for powerful movie-like dynamic graphic visualization of the gantry, table, MLC, room, lasers, 3D observer placement and beam centerline all as a function of MU or time, for analysis of the motions before requiring actual linac time. Results: We used the tool to generate and deliver extended SAD “virtual isocenter” trajectories of various shapes such as parameterized circles and ellipses. We also demonstrated use of the tool in generating linac couch motions that simulate respiratory motion using analytical parameterized functions. Conclusion: The SAGE tool is a valuable resource to experiment with families of complex geometric trajectories for a TrueBeam Linac. It makes Developer Mode more accessible as a vehicle to quickly translate research ideas into machine readable scripts without programming knowledge. As an open source initiative, it also enables researcher collaboration on future developments. I am a full time employee at Varian Medical Systems, Palo Alto, California.

43 PARTICLE ACCELERATORS↗
DOE OSTI2022

Computational Estimation by Scientific Data Mining with Classical Methods to Automate Learning Strategies of Scientists

Experimental results are often plotted as 2-dimensional graphical plots (aka graphs) in scientific domains depicting dependent versus independent variables to aid visual analysis of processes. Repeatedly performing laboratory experiments consumes significant time and resources, motivating the need for computational estimation. The goals are to estimate the graph obtained in an experiment given its input conditions, and to estimate the conditions that would lead to a desired graph. Existing estimation approaches often do not meet accuracy and efficiency needs of targeted applications. We develop a computational estimation approach called AutoDomainMine that integrates clustering and classification over complex scientific data in a framework so as to automate classical learning methods of scientists. Knowledge discovered thereby from a database of existing experiments serves as the basis for estimation. Challenges include preserving domain semantics in clustering, finding matching strategies in classification, striking a good balance between elaboration and conciseness while displaying estimation results based on needs of targeted users, and deriving objective measures to capture subjective user interests. These and other challenges are addressed in this work. The AutoDomainMine approach is used to build a computational estimation system, rigorously evaluated with real data in Materials Science. Our evaluation confirms that AutoDomainMine provides desired accuracy and efficiency in computational estimation. It is extendable to other science and engineering domains as proved by adaptation of its sub-processes within fields such as Bioinformatics and Nanotechnology.

Computer Science↗
DOE OSTI2021

AtomSets as a hierarchical transfer learning framework for small and large materials datasets

Abstract Predicting properties from a material’s composition or structure is of great interest for materials design. Deep learning has recently garnered considerable interest in materials predictive tasks with low model errors when dealing with large materials data. However, deep learning models suffer in the small data regime that is common in materials science. Here we develop the AtomSets framework, which utilizes universal compositional and structural descriptors extracted from pre-trained graph network deep learning models with standard multi-layer perceptrons to achieve consistently high model accuracy for both small compositional data (<400) and large structural data (>130,000). The AtomSets models show lower errors than the graph network models at small data limits and other non-deep-learning models at large data limits. They also transfer better in a simulated materials discovery process where the targeted materials have property values out of the training data limits. The models require minimal domain knowledge inputs and are free from feature engineering. The presented AtomSets model framework can potentially accelerate machine learning-assisted materials design and discovery with less data restriction.

Chen, Chi (ORCID:0000000180087043)↗
DOE OSTI2021

Optimizing FPGA-based Accelerator Design for Large-Scale Molecular Similarity Search (Special Session Paper)

Molecular similarity search has been widely used in drug discovery to rapidly identify structurally similar compounds from large molecular databases. With the increasing size of chemical libraries, there is growing interest in the efficient ac- celeration of large-scale similarity search. Existing works mainly focus on CPU and GPU to accelerate the computation of Tatimoto coefficient in measuring the pairwise similarity between different molecular fingerprints. In this paper, we propose and optimize an FPGA-based accelerator design on exhaustive and approximate search algorithms. On exhaustive search using BitBound & fold- ing, we analyze the similarity cutoff and folding level relationship with search speedup and accuracy, and propose a scalable on- the-fly query engine on FPGAs to reduce the resource utilization and pipeline interval. We achieve a 450 million compounds-per- second processing throughput for a single query engine. On approximate search using hierarchical navigable small world (HNSW), a popular algorithm with high recall and query speed, we propose an FPGA-based graph traversal engine to utilize high throughput register array based priority queue and fine- grained distance calculation engine to increase the processing capability. Experimental results show that the proposed FPGA- based HNSW implementation achieves a 35× speedup than existing works on CPU. To the best of our knowledge, our FPGA- based implementation is the first attempt to accelerate molecular similarity search on FPGA and has the highest performance among existing approaches.

Peng, Hongwu↗
DOE OSTI2017

Publication and Retrieval of Computational Chemical-Physical Data Via the Semantic Web. Final Technical Report

This research showed the feasibility of applying the concepts of the Semantic Web to Computation Chemistry. We have created the first web portal (www.chemsem.com) that allows data created in the calculations of quantum chemistry, and other such chemistry calculations to be placed on the web in a way that makes the data accessible to scientists in a semantic form never before possible. The semantic web nature of the portal allows data to be searched, found, and used as an advance over the usual approach of a relational database. The semantic data on our portal has the nature of a Giant Global Graph (GGG) that can be easily merged with related data and searched globally via a SPARQL Protocol and RDF Query Language (SPARQL) that makes global searches for data easier than with traditional methods. Our Semantic Web Portal requires that the data be understood by a computer and hence defined by an ontology (vocabulary). This ontology is used by the computer in understanding the data. We have created such an ontology for computational chemistry (purl.org/gc) that encapsulates a broad knowledge of the field of computational chemistry. We refer to this ontology as the Gainesville Core. While it is perhaps the first ontology for computational chemistry and is used by our portal, it is only a start of what must be a long multi-partner effort to define computational chemistry. In conjunction with the above efforts we have defined a new potential file standard (Common Standard for eXchange – CSX for computational chemistry data). This CSX file is the precursor of data in the Resource Description Framework (RDF) form that the semantic web requires. Our portal translates CSX files (as well as other computational chemistry data files) into RDF files that are part of the graph database that the semantic web employs. We propose a CSX file as a convenient way to encapsulate computational chemistry data.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2017

Disconnected Diagrams in Lattice QCD

In this work, we present state-of-the-art numerical methods and their applications for computing a particular class of observables using lattice quantum chromodynamics (Lattice QCD), a discretized version of the fundamental theory of quarks and gluons. These observables require calculating so called \disconnected diagrams" and are important for understanding many aspects of hadron structure, such as the strange content of the proton. We begin by introducing the reader to the key concepts of Lattice QCD and rigorously define the meaning of disconnected diagrams through an example of the Wick contractions of the nucleon. Subsequently, the calculation of observables requiring disconnected diagrams is posed as the computationally challenging problem of finding the trace of the inverse of an incredibly large, sparse matrix. This is followed by a brief primer of numerical sparse matrix techniques that overviews broadly used methods in Lattice QCD and builds the background for the novel algorithm presented in this work. We then introduce singular value deflation as a method to improve convergence of trace estimation and analyze its effects on matrices from a variety of fields, including chemical transport modeling, magnetohydrodynamics, and QCD. Finally, we apply this method to compute observables such as the strange axial charge of the proton and strange sigma terms in light nuclei. The work in this thesis is innovative for four reasons. First, we analyze the effects of deflation with a model that makes qualitative predictions about its effectiveness, taking only the singular value spectrum as input, and compare deflated variance with different types of trace estimator noise. Second, the synergy between probing methods and deflation is investigated both experimentally and theoretically. Third, we use the synergistic combination of deflation and a graph coloring algorithm known as hierarchical probing to conduct a lattice calculation of light disconnected matrix elements of the nucleon at two different values of the lattice spacing. Finally, we employ these algorithms to do a high-precision study of strange sigma terms in light nuclei; to our knowledge this is the first calculation of its kind from Lattice QCD.

Gambhir, Arjun↗
DOE OSTI2018

Visual Analytics of Heterogeneous Data Using Hypergraph Learning

For real-world learning tasks (e.g., classification), graph-based models are commonly used to fuse the information distributed in diverse data sources, which can be heterogeneous, redundant, and incomplete. Thesemodels represent the relations in different datasets as pairwise links. However, these links cannot deal with high-order relations which connect multiple objects (e.g., in public health datasets, more than two patient groups admitted by the same hospital in 2014). In this article, we propose a visual analytics approach forthe classification on heterogeneous datasets using the hypergraph model. The hypergraph is an extension to traditional graphs in which a hyperedge connects multiple vertices instead of just two. We model various high-order relations in heterogeneous datasets as hyperedges and fuse different datasets with a unified hypergraph structure. We use the hypergraph learning algorithm for predicting missing labels in the datasets.To allow users to inject their domain knowledge into the model-learning process, we augment the traditional learning algorithm in a number of ways. In addition, we also propose a set of visualizations which enable the user to construct the hypergraph structure and the parameters of the learning model interactively during the analysis. Here, we demonstrate the capability of our approach via two real-world cases.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2013

Effect of DC voltage pulses on memristor behavior.

Current knowledge of memristor behavior is limited to a few physical models of which little comprehensive data collection has taken place. The purpose of this research is to collect data in search of exploitable memristor behavior by designing and implementing tests on a HP Labs Rev2 Memristor Test Board. The results are then graphed in their optimal format for conceptualizing behavioral patterns. This series of experiments has concluded the existence of an additional memristor state affecting the behavior of memristors when pulsed with positively polarized DC voltages. This effect has been observed across multiple memristors and data sets. The following pages outline the process that led to the hypothetical existence and eventual proof of this additional state of memristor behavior.

Evans, Brian R.↗
DOE OSTI2017

Oak Ridge Graph Analytics for Medical Innovation (ORiGAMI)

In this era of data-driven decisions and discovery where Big Data is producing Bigger Data, data scientists at the Oak Ridge National Laboratory are leveraging unique leadership infrastructure (e.g., Urika XA and Urika GD appliances) to develop scalable algorithms for semantic, logical and statistical reasoning with Big Data (i.e., data stored in databases as well as unstructured data in documents). ORiGAMI is a next-generation knowledge-discovery framework that is: (a) knowledge nurturing (i.e., evolves seamlessly with newer knowledge and data), (b) smart and curious (i.e. using information-foraging and reasoning algorithms to digest content) and (c) synergistic (i.e., interfaces computers with what they do best to help subject-matter-experts do their best. ORiGAMI has been demonstrated using the National Library of Medicine's SEMANTIC MEDLINE (archive of medical knowledge since 1994).

Roberts, LarryW.↗
DOE OSTI2016

Oak Ridge Graph Analytics for Medical Innovation (ORiGAMI)

In this era of data-driven decisions and discovery where Big Data is producing Bigger Data, data scientists at the Oak Ridge National Laboratory are leveraging unique leadership infrastructure (e.g., Urika XA and Urika GD appliances) to develop scalable algorithms for semantic, logical and statistical reasoning with Big Data (i.e., data stored in databases as well as unstructured data in documents). ORiGAMI is a next-generation knowledge-discovery framework that is: (a) knowledge nurturing (i.e., evolves seamlessly with newer knowledge and data), (b) smart and curious (i.e. using information-foraging and reasoning algorithms to digest content) and (c) synergistic (i.e., interfaces computers with what they do best to help subject-matter-experts do their best. ORiGAMI has been demonstrated using the National Library of Medicine's SEMANTIC MEDLINE (archive of medical knowledge since 1994).

Roberts, Larry W.↗