Search NASA⌕ Search

SEARCH · Search NASA

Results for “adaptive 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 1,099 records · Page 61

Multisensor Agile Adaptive Sampling of Convective Storms Driven by Real-time Analytics

Convective storms vertically transport water vapor and condensate from Earth’s surface to the upper troposphere. Life on Earth is fundamentally linked to this transport which determines the hydrological cycle, and the intensity of severe weather responsible for the destruction of life and property. Despite advances in high-resolution modeling and better observational capabilities, the scientific community continues to be confronted with knowledge gaps about convective storms that limit our predictive capabilities. The ongoing developments in the high-resolution Energy Exascale Earth System Model (E3SM), large eddy simulations, and AI-based analytics to evaluate uncertainties are expected to provide a comprehensive framework for new scientific discovery. The model-experiment (MODEX) approach suggests that the aforementioned advancements in model development and AI-based inference techniques should be complemented by similar advancements in the experimental (observational) side so that the former does not outstrip the ability of the latter to provide meaningful constraints. What are the recent advancements in observations that will provide the necessary leap forward in improving our predictive capabilities? To address this question, we propose a new experimental paradigm called Multisensor Agile Adaptive Sampling (MAAS) that capitalizes on advancements in communications (5G), computational resources (edge/fog computing), sensor capabilities, and machine learning (ML) and AI techniques (Kollias et al., 2020). The MAAS framework allows for the collection of higher spatiotemporal resolution and quality observations of convective storms than is traditionally possible. The MAAS framework is scalable and applicable to atmospheric observatories such as those operated by the Department of Energy (DoE) Atmospheric Radiation Measurement (ARM) facility.

54 ENVIRONMENTAL SCIENCES↗

Developing and testing capabilities for simulating cases with heterogeneous land/water surfaces in a novel atmospheric large eddy simulation code

Large eddy simulations (LES) are the primary computational tool used to simulate high Reynolds number three-dimensional turbulent flows. In the context of earth system sciences, particularly atmospheric science, LES are uniquely able to resolve the scales of atmospheric motion that are key for building process-level understanding of boundary layer turbulence, atmosphere-surface interaction, clouds, and cloud-aerosol-chemistry interaction, and are a core limited-area modeling capability. Increasing demands are being placed on LES code bases as growing high performance computing resources allow LES to address a wider range of scientific problems. In addition, LES are emerging as a source of high-quality machine learning training data. These demands necessitate an agile and extensible code base that allows the model to quickly adapt to emergent needs. However, LES have largely relied on legacy Fortran code bases that lack flexibility. A new, Python-based LES capability called Predicting INteractions of Aerosol and Clouds in Large Eddy Simulation (PINACLES) has been developed as part of the Department of Energy’s Earth System Model Development (ESMD) program area’s Enabling Aerosol-cloud interactions at Global convection-permitting scalES (EAGLES) project. PINACLES was developed from the ground up with a philosophy of maximizing scientific throughput, by attempting to optimize for both model throughput and software extensibility. The initial development of PINACLES delivered a state-of-the-art idealized LES capability solving the non-hydrostatic anelastic equations of motion with doubly periodic boundary conditions and idealized homogenous surface boundary conditions. Here we provide a final report on the outcomes of a fiscal year 2021 Seed Laboratory Directed Research Project that extended PINACLES in two key ways. First, PINACLES was coupled to a state-of-the-art land surface model enabling it to simulate spatially inhomogeneous land-atmosphere interactions that are known to control key atmospheric processes. Second, the dynamical core of PINACLES was modified to permit non-periodic boundary conditions. This model enhancement enables simulation of realistic cases with boundary conditions prescribed from atmospheric reanalysis and enables nested simulations conducted on a hierarchy of computational domains with increasing resolution. Together, these extensions to PINACLES make it a formidable modeling capability and expand its potential application to diverse components of DOE’s atmospheric science portfolio.

42 ENGINEERING↗

The effect of simultaneous exposure on the attention selection and integration of segments and lexical tones by Urdu-Cantonese bilingual speakers

In the perceptual learning of lexical tones, an automatic and robust attention-to-phonology system enables native tonal listeners to adapt to acoustically non-optimal speech, such as phonetic conflicts in daily communications. Previous tone research reveals that non-native listeners who do not linguistically employ lexical tones in their mother tongue may find it challenging to attend to the tonal dimension or integrate it with the segmental features. However, it is unknown whether the attentional interference initially caused by a maternal attentional system would continue influencing the non-optimal tone perception for simultaneous bilingual teenagers. From an endpoint in the age of language acquisition, we investigate whether the tone-specific attention mechanism developed by the Urdu-Cantonese simultaneous bilinguals is automatic enough to assist them in adapting to a phonetically-conflicting environment. Three groups of teenagers engaged in a four-condition ABX task: Urdu-Cantonese simultaneous bilinguals, Cantonese native listeners, and Urdu-speaking, late learners of Cantonese. The results showed that although the simultaneous bilinguals could phonologically process Cantonese tones in a Cantonese-like way under a conflict-free listening condition, they still failed in adapting to the phonetic conflicts, especially the segment-induced ones. It thus demonstrated that the simultaneous exposure and years of regular education in Hong Kong local schools still could not automatically guarantee simultaneous bilingual processing of Cantonese tones. In interpreting the findings, it hypothesized that, except for simultaneous exposure, the development of a tone-specific attention mechanism is also likely to be L1-inhibitory, tone experience-driven, and language-specific for simultaneous bilinguals.

Ning, Jinghong↗

Testing Classical Properties from Quantum Data

Many properties of Boolean functions can be tested far more efficiently than the function itself can be learned. However, this dramatic advantage often disappears when testers are limited to random samples of ƒ instead of adaptively chosen queries to f. In this work we investigate the quantum version of this restriction: quantum algorithms that test properties of a Boolean function f solely from copies of either the function state |ƒ⟩ ∝ ∑ x |x, ƒ(x)⟩ or the phase state |(-1) ƒ ⟩ ∝ ∑ x (-1) ƒ(x) |x⟩. For monotonicity, symmetry, and triangle-freeness, we show passive quantum testers are unboundedly or super-polynomially better than their classical passive testing counterparts. They are competitive with classic query -based testers in each case. Our new testers use techniques beyond quantum Fourier sampling, and it turns out this is necessary: we show a certain class of bent functions can be tested from 𝒪(1) function states but has a sample complexity lower bound of 2 Ω(n) for any tester relying exclusively on Fourier and classical samples. Our passive quantum testers are competitive with classical query -based testers, but this isn't universal: we exhibit a testing problem that can be solved from 𝒪(1) classical queries but requires Ω(2 n/2 ) function state copies. The Forrelation problem provides a separation of the same magnitude in the opposite direction, so we conclude that quantum data and classical queries are "maximally incomparable" resources for testing. We also begin the study of lower bounds for testing from quantum data. For quantum monotonicity testing, we prove that the ensembles of [Goldreich et al., 2000; Black, 2024], which give exponential lower bounds for classical sample-based testing, do not yield any nontrivial lower bounds for testing from quantum data. New insights specific to quantum data will be required for proving copy complexity lower bounds for testing in this model.

Boolean Functions↗

Design and performance of a large vocabulary discrete word recognition system. Volume 1: Technical report

The development, construction, and test of a 100-word vocabulary near real time word recognition system are reported. Included are reasonable replacement of any one or all 100 words in the vocabulary, rapid learning of a new speaker, storage and retrieval of training sets, verbal or manual single word deletion, continuous adaptation with verbal or manual error correction, on-line verification of vocabulary as spoken, system modes selectable via verification display keyboard, relationship of classified word to neighboring word, and a versatile input/output interface to accommodate a variety of applications.

Source record↗

Neural network applications in telecommunications

Neural network capabilities include automatic and organized handling of complex information, quick adaptation to continuously changing environments, nonlinear modeling, and parallel implementation. This viewgraph presentation presents Bellcore work on applications, learning chip computational function, learning system block diagram, neural network equalization, broadband access control, calling-card fraud detection, software reliability prediction, and conclusions.

Alspector, Joshua↗

Upper Torso Control for HOAP-2 Using Neural Networks

Humanoid robots have similar physical builds and motion patterns as humans. Not only does this provide a suitable operating environment for the humanoid but it also opens up many research doors on how humans function. The overall objective is replacing humans operating in unsafe environments. A first target application is assembly of structures for future lunar-planetary bases. The initial development platform is a Fujitsu HOAP-2 humanoid robot. The goal for the project is to demonstrate the capability of a HOAP-2 to autonomously construct a cubic frame using provided tubes and joints. This task will require the robot to identify several items, pick them up, transport them to the build location, then properly assemble the structure. The ability to grasp and assemble the pieces will require improved motor control and the addition of tactile feedback sensors. In recent years, learning-based control is becoming more and more popular; for implementing this method we will be using the Adaptive Neural Fuzzy Inference System (ANFIS). When using neural networks for control, no complex models of the system must be constructed in advance-only input/output relationships are required to model the system.

Sandoval, Steven P.↗

Multi-Stage System for Automatic Target Recognition

A multi-stage automated target recognition (ATR) system has been designed to perform computer vision tasks with adequate proficiency in mimicking human vision. The system is able to detect, identify, and track targets of interest. Potential regions of interest (ROIs) are first identified by the detection stage using an Optimum Trade-off Maximum Average Correlation Height (OT-MACH) filter combined with a wavelet transform. False positives are then eliminated by the verification stage using feature extraction methods in conjunction with neural networks. Feature extraction transforms the ROIs using filtering and binning algorithms to create feature vectors. A feedforward back-propagation neural network (NN) is then trained to classify each feature vector and to remove false positives. The system parameter optimizations process has been developed to adapt to various targets and datasets. The objective was to design an efficient computer vision system that can learn to detect multiple targets in large images with unknown backgrounds. Because the target size is small relative to the image size in this problem, there are many regions of the image that could potentially contain the target. A cursory analysis of every region can be computationally efficient, but may yield too many false positives. On the other hand, a detailed analysis of every region can yield better results, but may be computationally inefficient. The multi-stage ATR system was designed to achieve an optimal balance between accuracy and computational efficiency by incorporating both models. The detection stage first identifies potential ROIs where the target may be present by performing a fast Fourier domain OT-MACH filter-based correlation. Because threshold for this stage is chosen with the goal of detecting all true positives, a number of false positives are also detected as ROIs. The verification stage then transforms the regions of interest into feature space, and eliminates false positives using an artificial neural network classifier. The multi-stage system allows tuning the detection sensitivity and the identification specificity individually in each stage. It is easier to achieve optimized ATR operation based on its specific goal. The test results show that the system was successful in substantially reducing the false positive rate when tested on a sonar and video image datasets.

Chao, Tien-Hsin↗

Safe Exploration: Sensorimotor Assessments for Early Extravehicular Activities

BACKGROUND: Artemis missions will require a new level of crew autonomy around periods of gravitational transition, where sensorimotor disturbances are at their highest. There is a need to define performance thresholds for key sensorimotor assessments that indicate when performance in early extravehicular activities (EVAs) might be impacted or unsafe. This panel presentation will discuss the development of sensorimotor assessments for determining crew preparedness of early EVAs by utilizing a novel portable sensorimotor disorientation analog and other spaceflight analogs. OVERVIEW: To define performance thresholds, a proposed set of sensorimotor assessment tasks must be validated under various spaceflight analogs. A Sensorimotor Disorientation Analog (SDA) was developed that could induce varying levels of disorientation through combined vestibular (galvanic vestibular stimulation (GVS)) and proprioceptive (weighted chest, ankles, and wrists) disruptions. The SDA was first pilot tested using subjective feedback from previously flown astronauts to determine the levels of disorientation that mimic motor performance immediately (R+0) and +24 hours (R+1) postflight. A second study was performed using healthy non-astronaut ground subjects to validate the SDA levels by comparing to astronaut postflight data. The validated SDA was utilized in a third study to map performance in the proposed set of sensorimotor assessment tasks to operational tasks. The assessment tasks were defined based on lessons learned from Apollo and subject matter experts (e.g., flight surgeons) to include the following: 1) aid in progressive adaptation to the novel gravitational environment; 2) provide opportunities to develop strategies to recover from off-nominal body positions; and 3) mimic operational tasks such that crew can self-assess their potential ability to complete their missions. This presentation will conclude with a discussion on future validation studies of the proposed assessment tasks using other spaceflight analogs such as centrifugation and gravity offload systems. DISCUSSION: Exploration class missions will require crew to be able to self-assess and treat their sensorimotor dysfunction after gravity transitions, and in off-nominal situations they may be required to perform provocative, challenging tasks soon after landing. This panel presentation will discuss current and ongoing research strategies to address the sensorimotor risk on safe exploration during Artemis missions.

Sarah Moudy↗

Updating a PWR Simulator in Python

A nuclear reactor is a complicated system that is comprised of many systems and components. In order to fully model a reactor, several approximations and correlations must be used for a real-time simulation. In 1984 as part of a doctoral dissertation at Massachusetts Institute of Technology (MIT), a FORTRAN 77 program was developed to simulate a four-loop, Westinghouse pressurized water reactor (PWR). The program was later called PRISM and extended to include a comprehensive set of features accessible in a DOS-based user interface. This present work seeks to update the original PRISM program for cross platform deployment using the Python language. The goal of this present work is to update and improve a teaching tool for both students and reactor operators. Originally, PRISM was created as part of a PhD project in FORTRAN 77, but the code no longer works readily on modern operating systems. Python was chosen to rewrite the program because of the ease of use of the Python modules including Kivy as well as the portability of the program to various operating systems. This project aims to develop desktop and mobile versions of the software, allowing the user flexibility in choice of platform. This simulator is scheduled to be deployed in time for a live demonstration at the 2016 Annual ANS meeting. The project aims to produce a reactor simulator that will showcase the dynamics of a PWR system in real-time under various conditions. Some potential scenarios include startup conditions, scram operations, or failure conditions such as a break in a coolant leg. The user will be able to induce various failure scenarios and the program will simulate how the reactor will behave. The simulation will include a suite of monitoring functions that will allow automatic systems to attempt to adapt to the changing conditions, e.g., control rod movement. This program will provide an educational resource for anyone learning about nuclear reactor kinetics or dynamics of a reactor system. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

CSA, the French LILW Disposal Facility: Taking Stock of the 30 years' Operating Experience - 20009

French Andra's currently operational LILW repository, Aube disposal facility, is almost 30 years old. From the beginning, the repository has benefited from the lessons learned from the first disposal facility, the CSM. It has nevertheless continued to evolve, to improve and to adapt, seeking to gain in efficiency, safety and cost optimization. This is a look at what has changed at the CSA since 1992 - and the effect these changes had on its operating practices, construction principles and the entire management organization. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Low Dose Radiation-Induced Genome and Epigenome Instability Symposium and Epigenetic Mechanisms, DNA Repair, and Chromatin Symposium at the EMS 2008 Annual Meeting - October 2008

The Low Dose Radiation Symposium thoughtfully addressed ionizing radiation non-mutational but transmissable alterations in surviving cells. Deregulation of epigenetic processes has been strongly implicated in carcinogenesis, and there is increasing realization that a significant fraction of non-targeted and adaptive mechanisms in response to ionizing radiation are likely to be epigenetic in nature. Much remains to be learned about how chromatin and epigenetic regulators affect responses to low doses of radiation, and how low dose radiation impacts other epigenetic processes. The Epigenetic Mechanisms Symposium focused on on epigenetic mechanisms and their interplay with DNA repair and chromatin changes. Addressing the fact that the most well understood mediators of epigenetic regulation are histone modifications and DNA methylation. Low levels of radiation can lead to changes in the methylation status of certain gene promoters and the expression of DNA methyltransferases, However, epigenetic regulation can also involve changes in higher order chromosome structure.

54 ENVIRONMENTAL SCIENCES↗

SU-F-I-12: Region-Specific Dictionary Learning for Low-Dose X-Ray CT Reconstruction

Purpose: Dictionary learning based method has attracted more and more attentions in low-dose CT due to the superior performance on suppressing noise and preserving structural details. Considering the structures and noise vary from region to region in one imaging object, we propose a region-specific dictionary learning method to improve the low-dose CT reconstruction. Methods: A set of normal-dose images was used for dictionary learning. Segmentations were performed on these images, so that the training patch sets corresponding to different regions can be extracted out. After that, region-specific dictionaries were learned from these training sets. For the low-dose CT reconstruction, a conventional reconstruction, such as filtered back-projection (FBP), was performed firstly, and then segmentation was followed to segment the image into different regions. Sparsity constraints of each region based on its dictionary were used as regularization terms. The regularization parameters were selected adaptively according to different regions. A low-dose human thorax dataset was used to evaluate the proposed method. The single dictionary based method was performed for comparison. Results: Since the lung region is very different from the other part of thorax, two dictionaries corresponding to lung region and the rest part of thorax respectively were learned to better express the structural details and avoid artifacts. With only one dictionary some artifact appeared in the body region caused by the spot atoms corresponding to the structures in the lung region. And also some structure in the lung regions cannot be recovered well by only one dictionary. The quantitative indices of the result by the proposed method were also improved a little compared to the single dictionary based method. Conclusion: Region-specific dictionary can make the dictionary more adaptive to different region characteristics, which is much desirable for enhancing the performance of dictionary learning based method.

60 APPLIED LIFE SCIENCES↗

Machine learning the density functional theory potential energy surface for the inorganic halide perovskite CsPbBr 3

The temperature and pressure dependence of structural phase transitions determine the structure-functionality relationships in many technologically important materials. Harmonic Hamiltonians have proven successful in predicting the vibrational properties of many materials. However, they are inadequate for modeling structural phase transitions in crystals with potential energy surfaces that are either strongly anharmonic or nonconvex with respect to collective atomic displacements or homogeneous strains. In this paper we develop a framework to express highly anharmonic first-principles potential energy surfaces as polynomials of collective cluster deformations. We further adapt the approach to a nonlinear extension of the cluster expansion formalism through the use of an artificial neural net model. Finally, the machine learning models are trained on a large database of first-principles calculations and are shown to reproduce the potential energy surface with low error.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

A framework to evaluate machine learning crystal stability predictions

The rapid adoption of machine learning in various scientific domains calls for the development of best practices and community agreed-upon benchmarking tasks and metrics. We present Matbench Discovery as an example evaluation framework for machine learning energy models, here applied as pre-filters to first-principles computed data in a high-throughput search for stable inorganic crystals. We address the disconnect between (1) thermodynamic stability and formation energy and (2) retrospective and prospective benchmarking for materials discovery. Alongside this paper, we publish a Python package to aid with future model submissions and a growing online leaderboard with adaptive user-defined weighting of various performance metrics allowing researchers to prioritize the metrics they value most. To answer the question of which machine learning methodology performs best at materials discovery, our initial release includes random forests, graph neural networks, one-shot predictors, iterative Bayesian optimizers and universal interatomic potentials. We highlight a misalignment between commonly used regression metrics and more task-relevant classification metrics for materials discovery. Accurate regressors are susceptible to unexpectedly high false-positive rates if those accurate predictions lie close to the decision boundary at 0 eV per atom above the convex hull. The benchmark results demonstrate that universal interatomic potentials have advanced sufficiently to effectively and cheaply pre-screen thermodynamic stable hypothetical materials in future expansions of high-throughput materials databases.

Riebesell, Janosh↗

Quo Vadis Payload Safety?

As we complete the preparations for the fourth Hubble Space Telescope (HST) servicing mission, we note an anniversary approaching: it was 30 years ago in July that the first HST payload safety review panel meeting was held. This, in turn, was just over a year after the very first payload safety review, a Phase 0 review for the Tracking and Data Relay Satellite and its Inertial Upper Stage, held in June of 1977. In adapting a process that had been used in the review and certification of earlier Skylab payloads, National Aeronautics and Space Administration (NASA) engineers sought to preserve the lessons learned in the development of technical payload safety requirements, while creating a new process that would serve the very different needs of the new space shuttle program. Their success in this undertaking is substantiated by the fact that this process and these requirements have proven to be remarkably robust, flexible, and adaptable. Furthermore, the payload safety process has, to date, served us well in the critical mission of safeguarding our astronauts, cosmonauts, and spaceflight participants. Both the technical requirements and their interpretation, as well as the associated process requirements have grown, evolved, been streamlined, and have been adapted to fit multiple programs, including the International Space Station (ISS) program, the Shuttle/Mir program, and most recently the United States Constellation program. From its earliest days, it was anticipated that the payload safety process would be international in scope, and so it has been. European Space Agency (ESA), Japan Aerospace Exploration Agency (JAXA), German Space Agency (DLR), Canadian Space Agency (CSA), Russian Space Agency (RSA), and many additional countries have flown payloads on both the space shuttle and on the ISS. Our close cooperation and long-term working relationships have culminated in the franchising of the payload safety review process itself to our partners in ESA, which in turn will serve as a roadmap for extending the franchise to other Partners.

Fodroci, Michael P.↗

MO-G-BRE-06: Metrics of Success: Measuring Participation and Attitudes Related to Near-Miss Incident Learning Systems

Purpose: Interest in incident learning systems (ILS) for improving safety and quality in radiation oncology is growing, as evidenced by the upcoming release of the national ILS. However, an institution implementing such a system would benefit from quantitative metrics to evaluate performance and impact. We developed metrics to measure volume of reporting, severity of reported incidents, and changes in staff attitudes over time from implementation of our institutional ILS. Methods: We analyzed 2023 incidents from our departmental ILS from 2/2012–2/2014. Incidents were prospectively assigned a near-miss severity index (NMSI) at multidisciplinary review to evaluate the potential for error ranging from 0 to 4 (no harm to critical). Total incidents reported, unique users reporting, and average NMSI were evaluated over time. Additionally, departmental safety attitudes were assessed through a 26 point survey adapted from the AHRQ Hospital Survey on Patient Safety Culture before, 12 months, and 24 months after implementation of the incident learning system. Results: Participation in the ILS increased as demonstrated by total reports (approximately 2.12 additional reports/month) and unique users reporting (0.51 additional users reporting/month). Also, the average NMSI of reports trended lower over time, significantly decreasing after 12 months of reporting (p<0.001) but with no significant change at months 18 or 24. In survey data significant improvements were noted in many dimensions, including perceived barriers to reporting incidents such as concern of embarrassment (37% to 18%; p=0.02) as well as knowledge of what incidents to report, how to report them, and confidence that these reports were used to improve safety processes. Conclusion: Over a two-year period, our departmental ILS was used more frequently, incidents became less severe, and staff confidence in the system improved. The metrics used here may be useful for other institutions seeking to create or evaluate their own incident learning systems.

61 RADIATION PROTECTION AND DOSIMETRY↗

Overview With Results and Lessons Learned of the X-43A Mach 10 Flight

This paper provides an overview of the final flight of the NASA X-43A project. The project consisted of three flights, two planned for Mach 7 and one for Mach 10. The third and final flight, November 16, 2004, was the first Mach 10 flight demonstration of an airframe-integrated, scramjet-powered, hypersonic vehicle. The goals and objectives for the project as well as those for the third flight are presented. The configuration of the Hyper-X stack including the X-43A, Hyper-X launch vehicle, and Hyper-X research vehicle adapter is discussed. The second flight of the X-43A was successfully conducted on March 27, 2004. Mission differences, vehicle modifications and lessons learned from the second flight as they applied to the third flight are also discussed. An overview of flight 3 results is presented.

Marshall, Laurie A.↗