Results for “standard library” — Page 5

NASA NTRS2015

Biomaterials Out of Thin Air: in Situ, On-Demand Printing of Advanced Biocomposites

Upmass is the single most significant limitation of our current space mission capability. Although biomaterials and biocomposites have mass, strength, flexibility, and self-healing properties that could significantly reduce upmass, their use is limited by the following drawbacks: Expensive, specific production. Many biomaterials can only be produced as part of significant support ecosystem; Inaccessible functional customization. The grain of wood, the porosity of bone, and so on are an integral part of the materials' desired mechanical properties, but are not deterministic when the material is naturally grown; Limited compositions. Most biomaterials (unlike metal, plastic, etc.) cannot be easily combined or modified to produce new materials. This project builds on recent advances in: Synthetic biology. Libraries of standardized genetic parts which can be used for controlled cellular material production, delivery, and binding; 3D printing. Commercial off-the-shelf components which can be used to make of a pico- to nanoliter cell deposition system; Tissue engineering. Proven cell-compatible support hydrogels and scaffolds can be modified to bind the deposited biomaterials of interest. Objectives: Feasibility and benefit analysis. Two mission contexts span the concept's scope (see below); Proof-of-concept demonstration. A simple grid of two proteins, fluorescent for easy detection, to validate the core technology concept; Proposed implementations for follow-on work. Avenues for future work on each core component (host cell, production control, material delivery, material binding, etc.); Complementary studies exploration. A survey of other emerging areas (in situ resource utilization, protein engineering, etc.) with the potential to multiply our technology's impact. Potential Impacts: This application could dramatically expand manufacturing capabilities on Earth and in space: In situ resource utilization. A far greater range of materials and products will be available from the limited palette offered by in situ resource extraction techniques; Reduced equipment and material upmass for off-Earth habitats. Ready- to-use highly specialized construction materials (radiation hardened, compressive/tensile, light or dense) from an extremely low starting mass; Structured biomaterial production. New ready-to-use macro, micro, and molecular manufacturing techniques for traditional biomaterials such as wood, bone and shell; New and novel biocomposite creation. The ability to create completely novel material composites from any base material that cells can be engineered to produce. Suggested Mission Contexts: ISS part manufacturing. A 'minimal working example' making a finished biomaterial part aboard the International Space Station; A long-term Mars habitat. 'Cradle-to-grave' use at a hypothetical Mars habitat, covering everything from tools to construction materials. Alternate Abstract: Imagine being able to print anything from tools and composite building materials to food and human tissues. Imagine being on Mars with the ability to replace any broken part, whether it's a part of your spacesuit, your habitat, or your own body. We propose a technique that would allow just that. By printing 3D arrays of cells engineered to secrete the necessary materials, the abundant in situ resources of atmosphere and regolith become organic, inorganic, or organic-inorganic composite materials. Such materials include novel, biologically derived materials not previously possible to fabricate.

habitats↗

NASA NTRS2015

Biomaterials Out of Thin Air: in Situ, On-demand Printing of Advanced Biocomposites

Upmass is the single most significant limitation of our current space mission capability. Although biomaterials and biocomposites have mass, strength, flexibility, and self-healing properties that could significantly reduce upmass, their use is limited by the following drawbacks: Expensive, specific production. Many biomaterials can only be produced as part of significant support ecosystem; Inaccessible functional customization. The grain of wood, the porosity of bone, and so on are an integral part of the materials' desired mechanical properties, but are not deterministic when the material is naturally grown; Limited compositions. Most biomaterials (unlike metal, plastic, etc.) cannot be easily combined or modified to produce new materials. This project builds on recent advances in: Synthetic biology. Libraries of standardized genetic parts which can be used for controlled cellular material production, delivery, and binding; 3D printing. Commercial off-the-shelf components which can be used to make of a pico- to nanoliter cell deposition system; Tissue engineering. Proven cell-compatible support hydrogels and scaffolds can be modified to bind the deposited biomaterials of interest. Objectives: Feasibility and benefit analysis. Two mission contexts span the concept's scope (see below); Proof-of-concept demonstration. A simple grid of two proteins, fluorescent for easy detection, to validate the core technology concept; Proposed implementations for follow-on work. Avenues for future work on each core component (host cell, production control, material delivery, material binding, etc.); Complementary studies exploration. A survey of other emerging areas (in situ resource utilization, protein engineering, etc.) with the potential to multiply our technology's impact. Potential Impacts: This application could dramatically expand manufacturing capabilities on Earth and in space: In situ resource utilization. A far greater range of materials and products will be available from the limited palette offered by in situ resource extraction techniques; Reduced equipment and material upmass for off-Earth habitats. Ready- to-use highly specialized construction materials (radiation hardened, compressive/tensile, light or dense) from an extremely low starting mass; Structured biomaterial production. New ready-to-use macro, micro, and molecular manufacturing techniques for traditional biomaterials such as wood, bone and shell; New and novel biocomposite creation. The ability to create completely novel material composites from any base material that cells can be engineered to produce. Suggested Mission Contexts: ISS part manufacturing. A 'minimal working example' making a finished biomaterial part aboard the International Space Station; A long-term Mars habitat. 'Cradle-to-grave' use at a hypothetical Mars habitat, covering everything from tools to construction materials. Alternate Abstract: Imagine being able to print anything from tools and composite building materials to food and human tissues. Imagine being on Mars with the ability to replace any broken part, whether it's a part of your spacesuit, your habitat, or your own body. We propose a technique that would allow just that. By printing 3D arrays of cells engineered to secrete the necessary materials, the abundant in situ resources of atmosphere and regolith become organic, inorganic, or organic-inorganic composite materials. Such materials include novel, biologically derived materials not previously possible to fabricate.

habitats↗

NASA NTRSDate not supplied

Aero-Engines AI - A Machine-Learning App for Aircraft Engine Concepts Assessment

Effective deployment of machine-learning (ML) models could drive a high level of efficiency in aircraft engine conceptual design. Aero-Engines AI is a user-friendly app that has been created to deploy trained machine-learning (ML) models to assess aircraft engine concepts. It was created using tkinter, a GUI (graphical user interface) module that is built into the standard Python library. Employing tkinter greatly facilitates the sharing of ML application as an executable file which can be run on Windows machines (without the need to have Python or any library installed). The app gets user input for a turbofan design, preprocesses the input data, and deploys trained ML models to predict turbofan thrust specific fuel consumption (TSFC), engine weight, core size, and turbomachinery stage-counts. The ML predictive models were built by employing supervised deep-learning and K-nearest neighbor regression algorithms to study patterns in an existing open-source database of production and research turbofan engines. They were trained, cross-validated, and tested in Keras, an open-source neural networks API (application programming interface) written in Python, with TensorFlow (Google open-source artificial intelligence library) serving as the backend engine. The smooth deployment of these ML models using the app shows that Aero-Engines AI is an easy-touse and a time-saving tool for aircraft engine design-space exploration during the conceptual design stage. Current version of the app focuses on the performance prediction of conventional turbofans. However, the scope of the app can easily be expanded to include other engine types (such as turboshaft and hybrid-electric systems) after their ML models are developed. Overall, the use of a machine-learning app for aircraft engine concept assessment represents a promising area of development in aircraft engine conceptual design.

machine learning↗

NASA NTRSDate not supplied

Aero-Engines AI - A Machine-Learning App for Aircraft Engine Concepts Assessment

Effective deployment of trained machine-learning models could drive a high level of efficiency in aircraft engine conceptual design. Aero-Engines AI is a Windows app that has been created to deploy trained machine-learning models to assess aircraft engine concepts. It was created using tkinter, a GUI (graphical user interface) module that is built into the standard Python library. Employing tkinter greatly facilitates the sharing of machine-learning application as an executable file which can be run on Windows machines (without the need to have Python or any library installed). Current version of the app focuses on the performance prediction of conventional turbofans. The app gets user input for a turbofan design, preprocesses the input data, and deploys trained machine-learning models to predict turbofan thrust specific fuel consumption (TSFC), engine weight, core size, and turbomachinery stage-counts. The machine-learning predictive models were built by employing supervised deep-learning algorithm to study patterns in an existing open-source database of production and research turbofan engines. They were trained, cross-validated, and tested in Keras, an open-source neural networks API (application programming interface) written in Python, with TensorFlow (Google open-source artificial intelligence library) serving as the backend engine. The smooth deployment of these machine-learning models using the app shows that Aero-Engines AI is an easy-to-use and a time-saving tool for aircraft engine design-space exploration during the conceptual design stage.

machine learning↗

NASA NTRS2023

Aero-Engines AI - A Machine-Learning App for Aircraft Engine Concepts Assessment

Effective deployment of machine-learning (ML) models could drive a high level of efficiency in aircraft engine conceptual design. Aero-Engines AI is a user-friendly app that has been created to deploy trained machine-learning (ML) models to assess aircraft engine concepts. It was created using tkinter, a GUI (graphical user interface) module that is built into the standard Python library. Employing tkinter greatly facilitates the sharing of ML application as an executable file which can be run on Windows machines (without the need to have Python or any library installed). The app gets user input for a turbofan design, preprocesses the input data, and deploys trained ML models to predict turbofan thrust specific fuel consumption (TSFC), engine weight, core size, and turbomachinery stage-counts. The ML predictive models were built by employing supervised deep-learning and K-nearest neighbor regression algorithms to study patterns in an existing open-source database of production and research turbofan engines. They were trained, cross-validated, and tested in Keras, an open-source neural networks API (application programming interface) written in Python, with TensorFlow (Google open-source artificial intelligence library) serving as the backend engine. The smooth deployment of these ML models using the app shows that Aero-Engines AI is an easy-touse and a time-saving tool for aircraft engine design-space exploration during the conceptual design stage. Current version of the app focuses on the performance prediction of conventional turbofans. However, the scope of the app can easily be easily expanded to include other engine types (such as turboshaft and hybrid-electric systems) after their ML models are developed. Overall, the use of a machine-learning app for aircraft engine concept assessment represents a promising area of development in aircraft engine conceptual design.

machine learning↗

DOE OSTI2025

Surrogate-driven Variance-based Sensitivity Analysis of Thermal Storage Tanks in Integrated Energy Systems

Sensitivity analysis and uncertainty quantification are essential steps for enhancing the accuracy of computational models by identifying and mitigating uncertainties. This study focuses on these steps for the Thermal Energy Delivery System at Idaho National Laboratory, specifically targeting the thermocline tank. Using a Modelica/Dymola simulation model, the study perturbed various design parameters and boundary conditions, including shape factor, porosity, outlet temperature, inlet mass flow rate, and system pressure, to predict and quantify uncertainty in the tank’s ax- ial temperature. A dataset of over 1,000 simulations was generated, and surrogate models were developed using the pyMAISE (Michigan Artificial Intelligence Standard Environment) library, which is an Automatic Machine Learning library for nuclear engineering applications. The optimal model, a feedforward neural network with two hidden layers, achieved an R2 score above 0.99 and a mean absolute error below 1 Kelvin. Sensitivity analyses using Sobol indices and Fourier amplitude sensitivity testing methods on this surrogate model revealed that the inlet mass flow rate at initial timestamps and porosity significantly impacts predicted temperatures across all sensors and time steps.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2024

Fine Temperature Grid Continuous Energy Cross Section Generation for Monte Carlo Analysis of Xe-100 Design

The standard “A Compact ENDF (ACE)” data libraries used by Monte Carlo based reactor physics codes calculations are provided by Los Alamos National Laboratory (LANL) with a temperature interval mostly of 300 K (e.g. 300 K, 600 K, 900 K) for the cross sections and between 100 K and 200 K for the thermal scattering libraries (TSL). However, some codes such as MCNP lack capability to perform on-the-fly temperature interpolation during simulation both for neutron and TSL cross-sections. To evaluate the impact related to Doppler broadening and spectrum shift associated with TSL changes, this paper explores the potential of adopting a temperature grid finer than the ones contained in the standard data libraries. A 50 K temperature grid was employed to quantify the error in neutronics calculations due to temperature grid resolution. This was achieved by comparing the results of this study (50 K temperature interval) against the results obtained with standard data libraries (>100 K temperature interval). While the adopted grid primarily relies on the ENDF/B-VII.1 library, for neutron cross-sections, it utilizes ENDF/B-VIII.0 library for TSL. The analyses confirmed that the accuracy of neutronics calculations is satisfactory when using a 50 K temperature grid. Notably, adopting a 50 K temperature grid, as opposed to standard libraries or coarser temperature grids, could lead to a difference of no more than a few hundred pcm in dk for both fresh fuel and burnt fuel. The most sensitive reaction type to the temperature grid was as expected identified as the capture cross-section of U-238.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

NASA NTRS2013

Flight Software Math Library

The flight software (FSW) math library is a collection of reusable math components that provides typical math utilities required by spacecraft flight software. These utilities are intended to increase flight software quality reusability and maintainability by providing a set of consistent, well-documented, and tested math utilities. This library only has dependencies on ANSI C, so it is easily ported. Prior to this library, each mission typically created its own math utilities using ideas/code from previous missions. Part of the reason for this is that math libraries can be written with different strategies in areas like error handling, parameters orders, naming conventions, etc. Changing the utilities for each mission introduces risks and costs. The obvious risks and costs are that the utilities must be coded and revalidated. The hidden risks and costs arise in miscommunication between engineers. These utilities must be understood by both the flight software engineers and other subsystem engineers (primarily guidance navigation and control). The FSW math library is part of a larger goal to produce a library of reusable Guidance Navigation and Control (GN&C) FSW components. A GN&C FSW library cannot be created unless a standardized math basis is created. This library solves the standardization problem by defining a common feature set and establishing policies for the library s design. This allows the libraries to be maintained with the same strategy used in its initial development, which supports a library of reusable GN&C FSW components. The FSW math library is written for an embedded software environment in C. This places restrictions on the language features that can be used by the library. Another advantage of the FSW math library is that it can be used in the FSW as well as other environments like the GN&C analyst s simulators. This helps communication between the teams because they can use the same utilities with the same feature set and syntax.

McComas, David↗

DOE OSTI2021

Quantum Chemistry Calculations for Metabolomics

A primary goal of metabolomics and exposomics studies is to fully characterize the small molecule composition of complex biological and environmental samples. However, despite advances in analytical technologies over the past two decades, the majority of small molecules in complex samples are not readily identifiable due to the immense structural and chemical diversi-ty present within the metabolome and exposome. Current gold-standard identification methods rely on reference libraries built using authentic chemical materials (“standards”), which are not available for most molecules. Computational quantum chemis-try methods, which can be used to calculate chemical properties that are then measured by analytical platforms, offer an alter-native route for building reference libraries, i.e., in silico libraries for “standards-free” identification. In this review, we cover the major roadblocks currently facing metabolomics and discuss applications where quantum chemistry calculations offer a solu-tion. Several successful examples for nuclear magnetic resonance spectroscopy, ion mobility spectrometry, infrared spectros-copy, and mass spectrometry methods are reviewed. Finally, we consider current best practices, sources of error, and provide an outlook for quantum chemistry calculations in metabolomics and exposomics studies. We expect this review will inspire researchers in the field of small molecule identification to accelerate adoption of in silico methods for generation of reference libraries and to add quantum chemistry calculations as another tool at their disposal to characterize complex samples.

Metabolism↗

DOE OSTI2022

Standardizing a Renewed Fission Product Yield Library and Related Covariances (Part II)

A complete library of independent fission product yields including covariance information was generated by a sequential Bayesian methodology whose prior information consisted of independent fission product yields and branching ratios of the possible decay modes reported in the recently released ENDF/B-VIII.0 nuclear data library. The Bayesian procedure consisted in the adjustment of independent fission product yields to match the set of stable and long-lived cumulative yields reported in the ENDF/B-VIII.0 nuclear data library which, in this work, were considered as legacy values together with their uncertainties. According to the χ 2 -metric, the results obtained by this methodology show a systematic improvement in the agreement with legacy reported cumulative yields for the entire adjusted independent fission product yield library.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

NASA NTRS1994

Robotic tape library system level testing at NSA: Present and planned

In the present of declining Defense budgets, increased pressure has been placed on the DOD to utilize Commercial Off the Shelf (COTS) solutions to incrementally solve a wide variety of our computer processing requirements. With the rapid growth in processing power, significant expansion of high performance networking, and the increased complexity of applications data sets, the requirement for high performance, large capacity, reliable and secure, and most of all affordable robotic tape storage libraries has greatly increased. Additionally, the migration to a heterogeneous, distributed computing environment has further complicated the problem. With today's open system compute servers approaching yesterday's supercomputer capabilities, the need for affordable, reliable secure Mass Storage Systems (MSS) has taken on an ever increasing importance to our processing center's ability to satisfy operational mission requirements. To that end, NSA has established an in-house capability to acquire, test, and evaluate COTS products. Its goal is to qualify a set of COTS MSS libraries, thereby achieving a modicum of standardization for robotic tape libraries which can satisfy our low, medium, and high performance file and volume serving requirements. In addition, NSA has established relations with other Government Agencies to complete this in-house effort and to maximize our research, testing, and evaluation work. While the preponderance of the effort is focused at the high end of the storage ladder, considerable effort will be extended this year and next at the server class or mid range storage systems.

Shields, Michael F.↗

DOE OSTI2024

LibERI—A portable and performant multi-GPU accelerated library for electron repulsion integrals via OpenMP offloading and standard language parallelism

A portable and performant graphics processing unit (GPU)-accelerated library for electron repulsion integral (ERI) evaluation, named LibERI, has been developed and implemented via directive-based (e.g., OpenMP and OpenACC) and standard language parallelism (e.g., Fortran DO CONCURRENT). Offloaded ERIs consist of integrals over low and high contraction s, p, and d functions using the rotated-axis and Rys quadrature methods. GPU codes are factorized based on previous developments with two layers of integral screening and quartet presorting. In this work, the density screening is moved to the GPU to enhance the computational efficacy for large molecular systems. Here, the L-shells in the Pople basis set are also separated into pure S and P shells to increase the ERI homogeneity and reduce atomic operations and the memory footprint. LibERI is compatible with any quantum chemistry drivers supporting the MolSSI Driver Interface. Benchmark calculations of LibERI interfaced with the GAMESS software package were carried out on various GPU architectures and molecular systems. The results show that the LibERI performance is comparable to other state-of-the-art GPU-accelerated codes (e.g., TeraChem and GMSHPC) and, in some cases, outperforms conventionally developed ERI CUDA kernels (e.g., QUICK) while fully maintaining portability.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2022

m-CUBES An efficient and portable implementation of multi-dimensional integration for gpus

The task of multi-dimensional numerical integration is frequently encountered in physics and other scientific fields, e.g., in modeling the effects of systematic uncertainties in physical systems and in Bayesian parameter estimation. Multi-dimensional integration is often time-prohibitive on CPUs. Efficient implementation on many-core architectures is challenging as the workload across the integration space cannot be predicted a priori. We propose m-Cubes, a novel implementation of the well-known Vegas algorithm for execution on GPUs. Vegas transforms integration variables followed by calculation of a Monte Carlo integral estimate using adaptive partitioning of the resulting space. m-Cubes improves performance on GPUs by maintaining relatively uniform workload across the processors. As a result, our optimized Cuda implementation for Nvidia GPUs outperforms parallelization approaches proposed in past literature. We further demonstrate the efficiency of m-Cubes by evaluating a six-dimensional integral from a cosmology application, achieving significant speedup and greater precision than the CUBA library's CPU implementation of VEGAS. We also evaluate m-Cubes on a standard integrand test suite. m-Cubes outperforms the serial implementations of the Cuba and GSL libraries by orders of magnitude speedup while maintaining comparable accuracy. Our approach yields a speedup of at least 10 when compared against publicly available Monte Carlo based GPU implementations. In summary, m-Cubes can solve integrals that are prohibitively expensive using standard libraries and custom implementations. A modern C++ interface header-only implementation makes m-Cubes portable, allowing its utilization in complicated pipelines with easy to define stateful integrals. Compatibility with non-Nvidia GPUs is achieved with our initial implementation of m-Cubes using the Kokkos framework.

Sakiotis, Ioannis↗

NASA NTRS2002

ISAIA: Interoperable Systems for Archival Information Access

The ISAIA project was originally proposed in 1999 as a successor to the informal AstroBrowse project. AstroBrowse, which provided a data location service for astronomical archives and catalogs, was a first step toward data system integration and interoperability. The goals of ISAIA were ambitious: '...To develop an interdisciplinary data location and integration service for space science. Building upon existing data services and communications protocols, this service will allow users to transparently query hundreds or thousands of WWW-based resources (catalogs, data, computational resources, bibliographic references, etc.) from a single interface. The service will collect responses from various resources and integrate them in a seamless fashion for display and manipulation by the user.' Funding was approved only for a one-year pilot study, a decision that in retrospect was wise given the rapid changes in information technology in the past few years and the emergence of the Virtual Observatory initiatives in the US and worldwide. Indeed, the ISAIA pilot study was influential in shaping the science goals, system design, metadata standards, and technology choices for the virtual observatory. The ISAIA pilot project also helped to cement working relationships among the NASA data centers, US ground-based observatories, and international data centers. The ISAIA project was formed as a collaborative effort between thirteen institutions that provided data to astronomers, space physicists, and planetary scientists. Among the fruits we ultimately hoped would come from this project would be a central site on the Web that any space scientist could use to efficiently locate existing data relevant to a particular scientific question. Furthermore, we hoped that the needed technology would be general enough to allow smaller, more-focused community within space science could use the same technologies and standards to provide more specialized services. A major challenge to searching for data across a broad community is that information that describe some data products are either not relevant to other data or not applicable in the same way. Some previous metadata standard development efforts (e.g., in the earth science and library communities) have produced standards that are very large and difficult to support. To address this problem, we studied how a standard may be divided into separable pieces. Data providers that wish to participate in interoperable searches can support only those parts of the standard that are relevant to them. We prototyped a top-level metadata standard that was small and applicable to all space science data.

Hanisch, Robert J.↗

DOE OSTI2022

Classification of Wetland Vegetation Based on NDVI Time Series from the HLS Dataset

Natural wetlands are intrinsically heterogeneous and typically composed of a mosaic of ecosystem patches with different vegetation types. Hydrological and biogeochemical processes in wetlands vary strongly among these ecosystem patches. To date, most remote sensing classification approaches for wetland vegetation either rely on coarse images that cannot capture the spatial variability of wetland vegetation or rely on very-high-resolution multi-spectral images that are detailed but very sporadic in time (less than once per year). This study aimed to use NDVI time series, generated from NASA’s HLS dataset, to classify vegetation patches. We demonstrate our approach at a temperate, coastal lake, estuarine marsh. To classify vegetation patches, a standard time series library of the four land-cover patch types was built from referencing specific locations that were identified as “pure” pixels. These were identified using a single-time high-resolution image. We calculated the distance between the HLS-NDVI time series at each pixel and the “pure”-pixel standards for each land-cover type. The resulting true-positive classified rate was >73% for all patch types other than water lily. The classification accuracy was higher in pixels of a more uniform composition. A set of vegetation maps was created for the years 2016 to 2020 at our research site to identify the vegetation changes at the site as it is affected by rapid water elevation increases in Lake Erie. Our results reveal how changes in water elevation have changed the patch distribution in significant ways, leading to the local extinction of cattail by 2019 and a continuous increase in the area cover of water lily patches.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2025

Simplifying activations with linear approximations in neural networks

A key step in Neural Networks is activation. Among the different types of activation functions, sigmoid, tanh, and others involve the usage of exponents for calculation. From a hardware perspective, exponential implementation implies the usage of Taylor series or repeated methods involving many addition, multiplication, and division steps, and as a result are power-hungry and consume many clock cycles. We implement a piecewise linear approximation of the sigmoid function as a replacement for standard sigmoid activation libraries. This approach provides a practical alternative by leveraging piecewise segmentation, which simplifies hardware implementation and improves computational efficiency. In this paper, we detail piecewise functions that can be implemented using linear approximations and their implications for overall model accuracy and performance gain. Our results show that for the DenseNet, ResNet, and GoogLeNet architectures, the piecewise linear approximation of the sigmoid function provides faster execution times compared to the standard TensorFlow sigmoid implementation while maintaining comparable accuracy. Specifically, for MNIST with DenseNet, accuracy reaches 99.91% (Piecewise) vs. 99.97% (Base) with up to 1.31x speedup in execution time. For CIFAR-10 with DenseNet, accuracy improves to 98.97% (Piecewise) vs. 99.40% (Base) while achieving 1.24x faster execution. Similarly, for CIFAR-100 with DenseNet, the accuracy is 97.93% (Piecewise) vs. 98.39% (Base), with a 1.18x execution time reduction. These results confirm the proposed method’s capability to efficiently process large-scale datasets and computationally demanding tasks, offering a practical means to accelerate deep learning models, including LSTMs, without compromising accuracy.

Activation function↗

DOE OSTI2020

Standardized Architecture for a Mega-Biobank Phenomic Library: The Million Veteran Program (MVP)

Electronic health records (EHRs) provide a wealth of data for phenotype development in population health studies, and researchers invest considerable time to curate data elements and validate disease definitions. The ability to reproduce well-defined phenotypes increases data quality, comparability of results and expedites research. In this paper, we present a standardized approach to organize and capture phenotype definitions, resulting in the creation of an open, online repository of phenotypes. This resource captures phenotype development, provenance and process from the Million Veteran Program, a national mega-biobank embedded in the Veterans Health Administration (VHA). To ensure that the repository is searchable, extendable, and sustainable, it is necessary to develop both a proper digital catalog architecture and underlying metadata infrastructure to enable effective management of the data fields required to define each phenotype. Our methods provide a resource for VHA investigators and a roadmap for researchers interested in standardizing their phenotype definitions to increase portability.

Knight, Kathryn↗

DOE OSTI2020

Standardized Architecture for a Mega-Biobank Phenomic Library: The Million Veteran Program (MVP)

Electronic health records (EHRs) provide a wealth of data for phenotype development in population health studies, and researchers invest considerable time to curate data elements and validate disease definitions. The ability to reproduce well-defined phenotypes increases data quality, comparability of results and expedites research. In this paper, we present a standardized approach to organize and capture phenotype definitions, resulting in the creation of an open, online repository of phenotypes. This resource captures phenotype development, provenance and process from the Million Veteran Program, a national mega-biobank embedded in the Veterans Health Administration (VHA). To ensure that the repository is searchable, extendable, and sustainable, it is necessary to develop both a proper digital catalog architecture and underlying metadata infrastructure to enable effective management of the data fields required to define each phenotype. Our methods provide a resource for VHA investigators and a roadmap for researchers interested in standardizing their phenotype definitions to increase portability.

Knight, Kathryn↗