Search NASA⌕ Search

SEARCH · Search NASA

Results for “Active learning”

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 811 records · Page 45
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Understanding the Inner-Workings of Language Models Through Representation Dissimilarity

We use model stitching to understand the internal representations of language models. Similar to vision models, we find that "more is better," and representations learned with more data and larger width can improve the performance of weaker models via stitching. We likewise find that certain architecture choices, using GeLU vs SoLU activation functions, influence the quality of learned representations. Finally, model stitching (as opposed to other model diagnostic methods, like mode connectivity) can localize the different generalization strategies of text classifiers under domain shift to certain hidden layers.

Brown, Davis R.↗
NASA NTRS2007

Sensor On-orbit Calibration and Characterization Using Spacecraft Maneuvers

Spacecraft flight operations often require activities that involve different kinds of maneuvers for orbital adjustments (pitch, yaw, and roll). Different maneuvers, when properly planned and scheduled, can also be applied to support and/or to perform on-board sensor calibration and characterization. This paper uses MODIS (Moderate Resolution Imaging Spectroradiometer) as an example to illustrate applications of spacecraft maneuvers for Earth-observing sensors on-orbit calibration and characterization. MODIS is one of the key instruments for NASA's Earth Observing System (EOS) currently operated on-board the EOS Terra and Aqua spacecraft launched in December 1999 and May 2002, respectively. Since their launch, both Terra and Aqua spacecraft have made a number of maneuvers, specially the yaw and roll maneuvers, to support the MODIS on-orbit calibration and characterization. For both Terra and Aqua MODIS, near-monthly spacecraft roll maneuvers are executed for lunar observations. These maneuvers are carefully scheduled so that the lunar phase angles are nearly identical for each sensor's lunar observations. The lunar observations are used to track MODIS reflective solar bands (RSB) calibration stability and to inter-compare Terra and Aqua MODIS RSB calibration consistency. To date, two sets of yaw maneuvers (each consists of two series of 8 consecutive yaws) by the Terra spacecraft and one set by the Aqua spacecraft have been performed to validate MODIS solar diffuser (SD) bi-directional reflectance factor (BRF) and to derive SD screen transmission. Terra spacecraft pitch maneuvers, first made on March 26, 2003 and the second on April 14, 2003 (with the Moon in the spacecraft nadir view), have been applied to characterize MODIS thermal emissive bands (TEB) response versus scan angle (RVS). This is particularly important since the pre-launch TEB RSV measurements made by the sensor vendor were not successful. Terra MODIS TEB RVS obtained from pitch maneuvers have been used in the current LIB calibration algorithm. Lunq observations from pitch maneuvers also provided information to cross-calibrate MODIS with other sensors (MISR and ASTER) on the same platform. We will provide a summary of MODIS maneuver activities and their applications for MODIS calibration and characterization. The results and lessons learned discussed in this paper from MODIS maneuver activities will provide useful insights into future spacecraft and sensor operation.

Xiong, X.↗
DOE OSTI2020

Analysis of MINERvA tunings of GENIE based on Hadron Multiplicity using Machine Learning

Scientists with the MINERvA neutrino interaction experiment have long studied distributions of hadrons (protons and pions) to improve the models for these reactions. Until now, most work has relied on easy to find, energetic, and well tracked hadrons. But in order to take full advantage of MINERνA data it is necessary to develop an efficient enough tool to recognize hadron tracks with lower energy, more transverse angle and more challenging patterns of activity. In this analysis, a Machine Learning tool based on Convolutional Neural Network and Transfer Learning has been developed to tackle that issue. It is found to be extremely capable of recognizing hadron activities especially when there are few and not very energetic hadrons, the accuracy is 76%. But it is also found that this network is very reluctant to classify too many hadrons. The classification is also affected by the presence of other particles in the event such photons, hadrons below a threshold or neutrons. This network is used to compare hadron multiplicity of data to the MC models and distinguish between different MC tunes. Some MINERνA tunes are explored in this analysis, but none of them described the data very well, especially not for all different hadrons multiplicities at once. Four combinations of those tunes were more successful, but none of them could be recognized as the best one. Nevertheless, this analysis develops an ML network that sets the basis of further exploration of hadron multiplicity as a tool to analyze MC tunes at MINERνA and beyond.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

FAIR AI models in high energy physics

Abstract The findable, accessible, interoperable, and reusable (FAIR) data principles provide a framework for examining, evaluating, and improving how data is shared to facilitate scientific discovery. Generalizing these principles to research software and other digital products is an active area of research. Machine learning models—algorithms that have been trained on data without being explicitly programmed—and more generally, artificial intelligence (AI) models, are an important target for this because of the ever-increasing pace with which AI is transforming scientific domains, such as experimental high energy physics (HEP). In this paper, we propose a practical definition of FAIR principles for AI models in HEP and describe a template for the application of these principles. We demonstrate the template’s use with an example AI model applied to HEP, in which a graph neural network is used to identify Higgs bosons decaying to two bottom quarks. We report on the robustness of this FAIR AI model, its portability across hardware architectures and software frameworks, and its interpretability.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2021

Training Spiking Neural Networks with Synaptic Plasticity under Integer Representation

Neuromorphic computing is emerging as a promising Beyond Moore computing paradigm that employs event-triggered computation and non-von Neumann hardware. Spike Timing Dependent Plasticity (STDP) is a well-known bio-inspired learning rule that relies on activities of locally connected neurons to adjust the weights of their respective synapses. In this work, we analyze a basic STDP rule and its sensitivity on the different hyperparameters for training spiking neural networks (SNNs) with supervision, customized for a neuromorphic hardware implementation with integer weights. We compare the classification performance on four UCI datasets (iris, wine, breast cancer and digits) that depict varying levels of complexity. We perform a search for optimal set of hyperparameters using both grid search and Bayesian optimization. Through the use of Bayesian optimization, we show the general trends in hyperparameter sensitivity in SNN classification problem. With the best sets of hyperparameters, we achieve accuracies comparable to some of the best performing SNNs on these four datasets. With a highly optimized supervised STDP rule we show that these accuracies can be achieved with just 20 epochs of training.

Kulkarni, Shruti↗
DOE OSTI2022

DFSynthesizer: Dataflow-based Synthesis of Spiking Neural Networks to Neuromorphic Hardware

Spiking Neural Networks (SNNs) are an emerging computation model that uses event-driven activation and bio-inspired learning algorithms. SNN-based machine learning programs are typically executed on tile-based neuromorphic hardware platforms, where each tile consists of a computation unit called a crossbar, which maps neurons and synapses of the program. However, synthesizing such programs on an off-the-shelf neuromorphic hardware is challenging. This is because of the inherent resource and latency limitations of the hardware, which impact both model performance, e.g., accuracy, and hardware performance, e.g., throughput. We propose DFSynthesizer, an end-to-end framework for synthesizing SNN-based machine learning programs to neuromorphic hardware. The proposed framework works in four steps. First, it analyzes a machine learning program and generates SNN workload using representative data. Second, it partitions the SNN workload and generates clusters that fit on crossbars of the target neuromorphic hardware. Third, it exploits the rich semantics of the Synchronous Dataflow Graph (SDFG) to represent a clustered SNN program, allowing for performance analysis in terms of key hardware constraints such as number of crossbars, dimension of each crossbar, buffer space on tiles, and tile communication bandwidth. Finally, it uses a novel scheduling algorithm to execute clusters on crossbars of the hardware, guaranteeing hardware performance. We evaluate DFSynthesizer with 10 commonly used machine learning programs. Our results demonstrate that DFSynthesizer provides a much tighter performance guarantee compared to current mapping approaches.

Computer Science↗
DOE OSTI2016

A novel procedure on next generation sequencing data analysis using text mining algorithm

Background: Next-generation sequencing (NGS) technologies have provided researchers with vast possibilities in various biological and biomedical research areas. Efficient data mining strategies are in high demand for large scale comparative and evolutional studies to be performed on the large amounts of data derived from NGS projects. Topic modeling is an active research field in machine learning and has been mainly used as an analytical tool to structure large textual corpora for data mining. Methods: We report a novel procedure to analyse NGS data using topic modeling. It consists of four major procedures: NGS data retrieval, preprocessing, topic modeling, and data mining using Latent Dirichlet Allocation (LDA) topic outputs. The NGS data set of the Salmonella enterica strains were used as a case study to show the workflow of this procedure. The perplexity measurement of the topic numbers and the convergence efficiencies of Gibbs sampling were calculated and discussed for achieving the best result from the proposed procedure. Results: The output topics by LDA algorithms could be treated as features of Salmonella strains to accurately describe the genetic diversity of fliC gene in various serotypes. The results of a two-way hierarchical clustering and data matrix analysis on LDA-derived matrices successfully classified Salmonella serotypes based on the NGS data. The implementation of topic modeling in NGS data analysis procedure provides a new way to elucidate genetic information from NGS data, and identify the gene-phenotype relationships and biomarkers, especially in the era of biological and medical big data. Conclusion: The implementation of topic modeling in NGS data analysis provides a new way to elucidate genetic information from NGS data, and identify the gene-phenotype relationships and biomarkers, especially in the era of biological and medical big data.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Little Big Horn College Health and Wellness Center 45 kilo-Watt (kW) Solar Array (Final Report)

The report summarizes the activities completed, results, and lessons learned in installing at 45 kW photovoltaic solar array on the roof and as awnings at the 25,000 square foot Health and Wellness Center on the campus of the Little Big Horn College (LBHC) in Crow Agency, Montana. The installation is expected to displace greater than 16% of the current electricity requirements at the Health and Wellness Center and save LBHC more than $6,600 in operating costs annually.

14 SOLAR ENERGY↗
DOE OSTI2025

Summary of Potential Incidents and Consequences from Carbon Dioxide Pipeline and Storage Systems Construction and Operation

This document provides a high-level summary of the potential human and environmental impacts associated with carbon dioxide (CO 2 ) pipeline transport, injection, and storage activities. It also reviews lessons learned from natural and industrial analogs of CO 2 storage as well as case studies of notable accidental CO 2 releases. The analysis is structured into several key sections, each addressing specific resource impacts and potential consequences.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS1981

Solar plages and the interpretation of stellar Ca 2 H and K line variations in late type dwarfs

Approximately a dozen late type dwarf stars have been observed by Wilson (1978) to undergo cyclic variability in Ca II H and K line emission which seems analogous to the solar activity cycle. What might be learned about these stars from solar analogies? The Ca II K index is estimated variation of the Sun viewed as a star, and compared with the observed range of Wilson's stellar observations. Results indicate trends of increasing relative variation H-K(max)/H-K(min) with later spectral type, due to decreasing dilutional contribution of residual photospheric flux to a 1 A band at line center, and of increasing relative variation with decreasing relative time of rise to maximum tau rise/tau reminiscent of the observed solar correlation of a quick rise to sunspot maximum with a strong cycle.

Cook, J. W.↗
NASA NTRS1986

Base pairing and base mis-pairing in nucleic acids

In recent years we have learned that DNA is conformationally active. It can exist in a number of different stable conformations including both right-handed and left-handed forms. Using single crystal X-ray diffraction analysis we are able to discover not only additional conformations of the nucleic acids but also different types of hydrogen bonded base-base interactions. Although Watson-Crick base pairings are the predominant type of interaction in double helical DNA, they are not the only types. Recently, we have been able to examine mismatching of guanine-thymine base pairs in left-handed Z-DNA at atomic resolution (1A). A minimum amount of distortion of the sugar phosphate backbone is found in the G x T pairing in which the bases are held together by two hydrogen bonds in the wobble pairing interaction. Because of the high resolution of the analysis we can visualize water molecules which fill in to accommodate the other hydrogen bonding positions in the bases which are not used in the base-base interactions. Studies on other DNA oligomers have revealed that other types of non-Watson-Crick hydrogen bonding interactions can occur. In the structure of a DNA octamer with the sequence d(GCGTACGC) complexed to an antibiotic triostin A, it was found that the two central AT base pairs are held together by Hoogsteen rather than Watson-Crick base pairs. Similarly, the G x C base pairs at the ends are also Hoogsteen rather than Watson-Crick pairing. Hoogsteen base pairs make a modified helix which is distinct from the Watson-Crick double helix.

Wang, A. H. J.↗
NASA NTRS2002

Visual Inference Programming

The goal of visual inference programming is to develop a software framework data analysis and to provide machine learning algorithms for inter-active data exploration and visualization. The topics include: 1) Intelligent Data Understanding (IDU) framework; 2) Challenge problems; 3) What's new here; 4) Framework features; 5) Wiring diagram; 6) Generated script; 7) Results of script; 8) Initial algorithms; 9) Independent Component Analysis for instrument diagnosis; 10) Output sensory mapping virtual joystick; 11) Output sensory mapping typing; 12) Closed-loop feedback mu-rhythm control; 13) Closed-loop training; 14) Data sources; and 15) Algorithms. This paper is in viewgraph form.

Wheeler, Kevin↗
NASA NTRS2006

Logistics Lessons Learned in NASA Space Flight

The Vision for Space Exploration sets out a number of goals, involving both strategic and tactical objectives. These include returning the Space Shuttle to flight, completing the International Space Station, and conducting human expeditions to the Moon by 2020. Each of these goals has profound logistics implications. In the consideration of these objectives,a need for a study on NASA logistics lessons learned was recognized. The study endeavors to identify both needs for space exploration and challenges in the development of past logistics architectures, as well as in the design of space systems. This study may also be appropriately applied as guidance in the development of an integrated logistics architecture for future human missions to the Moon and Mars. This report first summarizes current logistics practices for the Space Shuttle Program (SSP) and the International Space Station (ISS) and examines the practices of manifesting, stowage, inventory tracking, waste disposal, and return logistics. The key findings of this examination are that while the current practices do have many positive aspects, there are also several shortcomings. These shortcomings include a high-level of excess complexity, redundancy of information/lack of a common database, and a large human-in-the-loop component. Later sections of this report describe the methodology and results of our work to systematically gather logistics lessons learned from past and current human spaceflight programs as well as validating these lessons through a survey of the opinions of current space logisticians. To consider the perspectives on logistics lessons, we searched several sources within NASA, including organizations with direct and indirect connections with the system flow in mission planning. We utilized crew debriefs, the John Commonsense lessons repository for the JSC Mission Operations Directorate, and the Skylab Lessons Learned. Additionally, we searched the public version of the Lessons Learned Information System (LLIS) and verified that we received the same result using the internal version of LLIS for our logistics lesson searches. In conducting the research, information from multiple databases was consolidated into a single spreadsheet of 300 lessons learned. Keywords were applied for the purpose of sorting and evaluation. Once the lessons had been compiled, an analysis of the resulting data was performed, first sorting it by keyword, then finding duplication and root cause, and finally sorting by root cause. The data was then distilled into the top 7 lessons learned across programs, centers, and activities.

Evans, William A.↗
NASA NTRS2011

Smackdown: Adventures in Simulation Standards and Interoperability

The paucity of existing employer-driven simulation education and the need for workers broadly trained in Modeling & Simulation (M&S) poses a critical need that the simulation community as a whole must address. This paper will describe how this need became an impetus for a new inter-university activity that allows students to learn about simulation by doing it. The event, called Smackdown, was demonstrated for the first time in April at the Spring Simulation Multi-conference. Smackdown is an adventure in international cooperation. Students and faculty took part from the US and Europe supported by IEEE/SISO standards, industry software and National Aeronautics and Space Administration (NASA) content of are supply mission to the Moon. The developers see Smackdown providing all participants with a memorable, interactive, problem-solving experience, which can contribute, importantly to the workforce of the future. This is part of the larger need to increase undergraduate education in simulation and could be a prime candidate for senior design projects.

Elfrey, Priscilla R.↗
NASA NTRS2013

Helicopter Field Testing of NASA's Autonomous Landing and Hazard Avoidance Technology (ALHAT) System fully integrated with the Morpheus Vertical Test Bed Avionics

The Autonomous Landing Hazard Avoidance Technology (ALHAT) Project was chartered to develop and mature to a Technology Readiness Level (TRL) of six an autonomous system combining guidance, navigation and control with real-time terrain sensing and recognition functions for crewed, cargo, and robotic planetary landing vehicles. The ALHAT System must be capable of identifying and avoiding surface hazards to enable a safe and accurate landing to within tens of meters of designated and certified landing sites anywhere on a planetary surface under any lighting conditions. This is accomplished with the core sensing functions of the ALHAT system: Terrain Relative Navigation (TRN), Hazard Detection and Avoidance (HDA), and Hazard Relative Navigation (HRN). The NASA plan for the ALHAT technology is to perform the TRL6 closed loop demonstration on the Morpheus Vertical Test Bed (VTB). The first Morpheus vehicle was lost in August of 2012 during free-flight testing at Kennedy Space Center (KSC), so the decision was made to perform a helicopter test of the integrated ALHAT System with the Morpheus avionics over the ALHAT planetary hazard field at KSC. The KSC helicopter tests included flight profiles approximating planetary approaches, with the entire ALHAT system interfaced with all appropriate Morpheus subsystems and operated in real-time. During these helicopter flights, the ALHAT system imaged the simulated lunar terrain constructed in FY2012 to support ALHAT/Morpheus testing at KSC. To the best of our knowledge, this represents the highest fidelity testing of a system of this kind to date. During this helicopter testing, two new Morpheus landers were under construction at the Johnson Space Center to support the objective of an integrated ALHAT/Morpheus free-flight demonstration. This paper provides an overview of this helicopter flight test activity, including results and lessons learned, and also provides an overview of recent integrated testing of ALHAT on the second Morpheus vehicle.

Rutishauser, David↗
NASA NTRS2013

Background Pressure Profiles for Sonic Boom Vehicle Testing in the NASA Glenn 8- by 6-Foot Supersonic Wind Tunnel

In an effort to identify test facilities that offer sonic boom measurement capabilities, an exploratory test program was initiated using wind tunnels at NASA research centers. The subject of this report is the sonic boom pressure rail data collected in the Glenn Research Center 8- by 6-Foot Supersonic Wind Tunnel. The purpose is to summarize the lessons learned based on the test activity, specifically relating to collecting sonic boom data which has a large amount of spatial pressure variation. The wind tunnel background pressure profiles are presented as well as data which demonstrated how both wind tunnel Mach number and model support-strut position affected the wind tunnel background pressure profile. Techniques were developed to mitigate these effects and are presented.

Castner, Raymond↗
NASA NTRS2013

Helicopter Field Testing of NASA's Autonomous Landing and Hazard Avoidance Technology (ALHAT) System fully Integrated with the Morpheus Vertical Test Bed Avionics

The Autonomous Landing and Hazard Avoidance Technology (ALHAT) Project was chartered to develop and mature to a Technology Readiness Level (TRL) of six an autonomous system combining guidance, navigation and control with real-time terrain sensing and recognition functions for crewed, cargo, and robotic planetary landing vehicles. The ALHAT System must be capable of identifying and avoiding surface hazards to enable a safe and accurate landing to within tens of meters of designated and certified landing sites anywhere on a planetary surface under any lighting conditions. This is accomplished with the core sensing functions of the ALHAT system: Terrain Relative Navigation (TRN), Hazard Detection and Avoidance (HDA), and Hazard Relative Navigation (HRN). The NASA plan for the ALHAT technology is to perform the TRL6 closed loop demonstration on the Morpheus Vertical Test Bed (VTB). The first Morpheus vehicle was lost in August of 2012 during free-flight testing at Kennedy Space Center (KSC), so the decision was made to perform a helicopter test of the integrated ALHAT System with the Morpheus avionics over the ALHAT planetary hazard field at KSC. The KSC helicopter tests included flight profiles approximating planetary approaches, with the entire ALHAT system interfaced with all appropriate Morpheus subsystems and operated in real-time. During these helicopter flights, the ALHAT system imaged the simulated lunar terrain constructed in FY2012 to support ALHAT/Morpheus testing at KSC. To the best of our knowledge, this represents the highest fidelity testing of a system of this kind to date. During this helicopter testing, two new Morpheus landers were under construction at the Johnson Space Center to support the objective of an integrated ALHAT/Morpheus free-flight demonstration. This paper provides an overview of this helicopter flight test activity, including results and lessons learned, and also provides an overview of recent integrated testing of ALHAT on the second Morpheus vehicle.

Epp, Chirold D.↗
NASA NTRSDate not supplied

NASA DEVELOP’s Approach to Co-Production and Collaborative STEM Engagement

The NASA DEVELOP Program conducts 50-60 projects each year with the goal of bringing the benefits of NASA Earth science to local decision-making challenges. DEVELOP, part of NASA’s Applied Sciences’ Capacity Building Program, connects end users with students, recent graduates, and early and transitioning career professionals through 10-week rapid feasibility studies. These projects are a collaboration between NASA, DEVELOP office host locations (universities and other federal installations), project partners (federal agencies, state and local governments, non-profit and for-profit organizations, and international organizations), and the project teams who conduct them. The projects take place under the guidance of science advisors from NASA, academia, and partner organizations, and introduces communities to new applications of NASA Earth observations data with the desired outcome of informed decision-making. This presentation will highlight the DEVELOP model of co-production and programmatic collaboration, lessons learned in partnering, and evaluation activities that look at the impact of the program’s efforts.

Capacity Building↗