Search NASA⌕ Search

Engineering topics

Walker, Nicholas

Publications and source records attributed to Walker, Nicholas.

9 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Structured information extraction from scientific text with large language models

Extracting structured knowledge from scientific text remains a challenging task for machine learning models. Here, we present a simple approach to joint named entity recognition and relation extraction and demonstrate how pretrained large language models (GPT-3, Llama-2) can be fine-tuned to extract useful records of complex scientific knowledge. We test three representative tasks in materials chemistry: linking dopants and host materials, cataloging metal-organic frameworks, and general composition/phase/morphology/application information extraction. Records are extracted from single sentences or entire paragraphs, and the output can be returned as simple English sentences or a more structured format such as a list of JSON objects. This approach represents a simple, accessible, and highly flexible route to obtaining large databases of structured specialized scientific knowledge extracted from research papers.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2023

Application of the variational autoencoder to detect the critical points of the anisotropic Ising model

We generalize the previous study on the application of variational autoencoders to the two-dimensional Ising model to a system with anisotropy. Due to the self-duality property of the system, the critical points can be located exactly for the entire range of anisotropic coupling. This presents an excellent test bed for the validity of using a variational autoencoder to characterize an anisotropic classical model. Furthermore, we reproduce the phase diagram for a wide range of anisotropic couplings and temperatures via a variational autoencoder without the explicit construction of an order parameter. Considering that the partition function of ($d$ + 1)-dimensional anisotropic models can be mapped to that of the $d$-dimensional quantum spin models, the present study provides numerical evidence that a variational autoencoder can be applied to analyze quantum systems via the quantum Monte Carlo method.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2022

Neural Network Solver for Small Quantum Clusters

Machine learning approaches have recently been applied to the study of various problems in physics. Most of these studies are focused on interpreting the data generated by conventional numerical methods or the data on an existing experimental database. An interesting question is whether it is possible to use a machine learning approach, in particular a neural network, for solving the many-body problem. In this paper, we present a neural network solver for the single impurity Anderson model, the paradigm of an interacting quantum problem in small clusters. We demonstrate that the neural-network-based solver provides quantitative accurate results for the spectral function as compared to the exact diagonalization method. This opens the possibility of utilizing the neural network approach as an impurity solver for other many-body numerical approaches, such as the dynamical mean field theory.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Deep learning on the 2-dimensional Ising model to extract the crossover region with a variational autoencoder

The 2-dimensional Ising model on a square lattice is investigated with a variational autoencoder in the non-vanishing field case for the purpose of extracting the crossover region between the ferromagnetic and paramagnetic phases. The encoded latent variable space is found to provide suitable metrics for tracking the order and disorder in the Ising configurations that extends to the extraction of a crossover region in a way that is consistent with expectations. The extracted results achieve an exceptional prediction for the critical point as well as agreement with previously published results on the configurational magnetizations of the model. The performance of this method provides encouragement for the use of machine learning to extract meaningful structural information from complex physical systems where little a priori data is available.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2018

Identifying structural changes with unsupervised machine learning methods

In this work, unsupervised machine learning methods are used to identify structural changes using the melting point transition in classical molecular dynamics simulations as an example application of the approach. Dimensionality reduction and clustering methods are applied to instantaneous radial distributions of atomic configurations from classical molecular dynamics simulations of metallic systems over a large temperature range. Principal component analysis is used to dramatically reduce the dimensionality of the feature space across the samples using an orthogonal linear transformation that preserves the statistical variance of the data under the condition that the new feature space is linearly independent. From there, k-means clustering is used to partition the samples into solid and liquid phases through a criterion motivated by the geometry of the reduced feature space of the samples, allowing for an estimation of the melting point transition. This pattern criterion is conceptually similar to how humans interpret the data but with far greater throughput, as the shapes of the radial distributions are different for each phase and easily distinguishable by humans. The transition temperature estimates derived from this machine learning approach produce comparable results to other methods on similarly small system sizes. These results show that machine learning approaches can be applied to structural changes in physical systems.

42 ENGINEERING↗
DOE OSTI2013

The International Linear Collider Technical Design Report - Volume 1: Executive Summary

The International Linear Collider Technical Design Report (TDR) describes in four volumes the physics case and the design of a 500 GeV centre-of-mass energy linear electron-positron collider based on superconducting radio-frequency technology using Niobium cavities as the accelerating structures. The accelerator can be extended to 1 TeV and also run as a Higgs factory at around 250 GeV and on the Z0 pole. A comprehensive value estimate of the accelerator is give, together with associated uncertainties. It is shown that no significant technical issues remain to be solved. Once a site is selected and the necessary site-dependent engineering is carried out, construction can begin immediately. The TDR also gives baseline documentation for two high-performance detectors that can share the ILC luminosity by being moved into and out of the beam line in a "push-pull" configuration. These detectors, ILD and SiD, are described in detail. They form the basis for a world-class experimental programme that promises to increase significantly our understanding of the fundamental processes that govern the evolution of the Universe.

43 PARTICLE ACCELERATORS↗
DOE OSTI2011

International Linear Collider-A Technical Progress Report

The International Linear Collider: A Technical Progress Report marks the halfway point towards the Global Design Effort fulfilling its mandate to follow up the ILC Reference Design Report with a more optimised Technical Design Report (TDR) by the end of 2012. The TDR will be based on much of the work reported here and will contain all the elements needed to propose the ILC to collaborating governments, including a technical design and implementation plan that are realistic and have been better optimised for performance, cost and risk. We are on track to develop detailed plans for the ILC, such that once results from the Large Hadron Collider (LHC) at CERN establish the main science goals and parameters of the next machine, we will be in good position to make a strong proposal for this new major global project in particle physics. The two overriding issues for the ILC R&D programme are to demonstrate that the technical requirements for the accelerator are achievable with practical technologies, and that the ambitious physics goals can be addressed by realistic ILC detectors. This GDE interim report documents the impressive progress on the accelerator technologies that can make the ILC a reality. It highlights results of the technological demonstrations that are giving the community increased confidence that we will be ready to proceed with an ILC project following the TDR. The companion detector and physics report document likewise demonstrates how detector designs can meet the ambitious and detailed physics goals set out by the ILC Steering Committee. LHC results will likely affect the requirements for the machine design and the detectors, and we are monitoring that very closely, intending to adapt our design as those results become available.

43 PARTICLE ACCELERATORS↗
DOE OSTI2011

International Linear Collider (ILC): A Technical Progress Report

In 2003, more than 2,700 scientists from around the world signed a published statement Understanding Matter, Energy, Space and Time: the Case for the Linear Collider [1-1]. The statement made clear the undisputed worldwide consensus that the next energy frontier machine after the Large Hadron Collider (LHC) should be an electron-positron linear collider with an initial centre-of-mass energy reach of 500 gigaelectronvolts (GeV), upgradable to 1 teraelectronvolt (TeV). In the same year, the International Committee for Future Accelerators (ICFA) commissioned the International Technology Review Panel (ITRP) to select from the then competing acceleration technologies a single technology that would be the basis for an international design effort towards a truly global project. The following year, 2004, the ITRP recommended to ICFA that the 1.3-gigahertz superconducting radiofrequency (SCRF) technology developed by the TESLA collaboration [1-2] should form the basis for the design of the International Linear Collider (ILC).

43 PARTICLE ACCELERATORS↗