SEARCH · Search NASA

Results for “learning efficiency”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 613 records · Page 34

All sources NASA NTRS DOE OSTI

DOE OSTI2022

FINETUNA: fine-tuning accelerated molecular simulations

Abstract Progress towards the energy breakthroughs needed to combat climate change can be significantly accelerated through the efficient simulation of atomistic systems. However, simulation techniques based on first principles, such as density functional theory (DFT), are limited in their practical use due to their high computational expense. Machine learning approaches have the potential to approximate DFT in a computationally efficient manner, which could dramatically increase the impact of computational simulations on real-world problems. However, they are limited by their accuracy and the cost of generating labeled data. Here, we present an online active learning framework for accelerating the simulation of atomic systems efficiently and accurately by incorporating prior physical information learned by large-scale pre-trained graph neural network models from the Open Catalyst Project. Accelerating these simulations enables useful data to be generated more cheaply, allowing better models to be trained and more atomistic systems to be screened. We also present a method of comparing local optimization techniques on the basis of both their speed and accuracy. Experiments on 30 benchmark adsorbate-catalyst systems show that our method of transfer learning to incorporate prior information from pre-trained models accelerates simulations by reducing the number of DFT calculations by 91%, while meeting an accuracy threshold of 0.02 eV 93% of the time. Finally, we demonstrate a technique for leveraging the interactive functionality built in to Vienna ab initio Simulation Package (VASP) to efficiently compute single point calculations within our online active learning framework without the significant startup costs. This allows VASP to work in tandem with our framework while requiring 75% fewer self-consistent cycles than conventional single point calculations. The online active learning implementation, and examples using the VASP interactive code, are available in the open source FINETUNA package on Github.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS2023

Efficient Calibration of Expensive Computational Models

Accounting for uncertainty when calibrating expensive computational models is a common challenge faced by scientists and engineers. Often Bayesian techniques are adopted to estimate a probability density function over the model parameters given noisy empirical data. The methods used to perform this type of probabilistic calibration are computationally prohibitive in that they require a large number of evaluations of the expensive model. In these cases, surrogate modeling -- that is, using a fast-to-evaluate, lower fidelity stand-in for the original computational model -- may be the only option to alleviate this computational burden. However, the upfront cost of generating training data to build a surrogate model can itself be expensive. As such, it is important to be judicious when selecting training points at which the full-fidelity model is evaluated. Here, an active learning approach is proposed that enables efficient selection of training points using approximate samples of the calibrated parameter probability density function. In this way, the training points can be concentrated in regions where the calibration algorithm requires high model accuracy.

active learning↗

NASA NTRS2004

Multimodality instrument for tissue characterization

A system with multimodality instrument for tissue identification includes a computer-controlled motor driven heuristic probe with a multisensory tip. For neurosurgical applications, the instrument is mounted on a stereotactic frame for the probe to penetrate the brain in a precisely controlled fashion. The resistance of the brain tissue being penetrated is continually monitored by a miniaturized strain gauge attached to the probe tip. Other modality sensors may be mounted near the probe tip to provide real-time tissue characterizations and the ability to detect the proximity of blood vessels, thus eliminating errors normally associated with registration of pre-operative scans, tissue swelling, elastic tissue deformation, human judgement, etc., and rendering surgical procedures safer, more accurate, and efficient. A neural network program adaptively learns the information on resistance and other characteristic features of normal brain tissue during the surgery and provides near real-time modeling. A fuzzy logic interface to the neural network program incorporates expert medical knowledge in the learning process. Identification of abnormal brain tissue is determined by the detection of change and comparison with previously learned models of abnormal brain tissues. The operation of the instrument is controlled through a user friendly graphical interface. Patient data is presented in a 3D stereographics display. Acoustic feedback of selected information may optionally be provided. Upon detection of the close proximity to blood vessels or abnormal brain tissue, the computer-controlled motor immediately stops probe penetration. The use of this system will make surgical procedures safer, more accurate, and more efficient. Other applications of this system include the detection, prognosis and treatment of breast cancer, prostate cancer, spinal diseases, and use in general exploratory surgery.

Mah, Robert W.↗

DOE OSTI2022

A Fast and Accurate Transient Stability Assessment Method Based on Deep Learning: WECC Case Study

Transient stability is one of the critical aspects of power system stability assessment. The increasing integration of inverter-based resources and the retirement of conventional synchronous generators result in the decreasing system inertia and growing complexity of system operating conditions. Using a few selected typical operating conditions cannot guarantee system transient stability in all operating conditions, and the time-domain simulation of all operating conditions requires tremendous time and is often infeasible. This paper proposes a more efficient transient stability assessment method based on deep learning. The binary search method is used to determine the critical clearing time (CCT) in creating training databased by time-domain simulation. This method is fast and accurate with 1 ms resolution. The buses whose CCTs are lower than 200 ms are considered critical buses. Buses close to each other are grouped based on their mutual admittance matrix to reduce the search space of the critical buses. This paper also proposes the generator feature normalization based on the physical model. Case study on the reduced 240-bus WECC system model demonstrates that the proposed method can predict CCT accurately and efficiently.

critical clearing time↗

DOE OSTI2022

An Intelligent Distributed Ledger Construction Algorithm for IoT

Blockchain is the next generation of secure data management that creates near-immutable decentralized storage. Secure cryptography created a niche for blockchain to provide alternatives to well-known security compromises. However, design bottlenecks with traditional blockchain data structures scale poorly with increased network usage and are extremely computation-intensive. This made the technology difficult to combine with limited devices, like those in Internet of Things networks. In protocols like IOTA, replacement of blockchain's linked-list queue processing with a lightweight dynamic ledger showed remarkable throughput performance increase. However, current stochastic algorithms for ledger construction suffer distinct trade-offs between efficiency and security. This work proposed a machine-learning approach with a multi-arm bandit that resolved these issues and was designed for auditing on limited devices. This algorithm was tested in a reinforcement-learning environment simulating the IOTA ledger's construction with a decision tree. This study showed through regret analysis and experimentation that this approach was secure against impulse manipulation attacks while remaining energy-efficient. Although the IOTA protocol was a pioneer for lightweight distributed ledgers, it is expected that future blockchain protocols will adopt techniques similar to those presented in this work.

multi-arm bandit↗

DOE OSTI2020

Learning compact physics‐aware delayed photocurrent models using dynamic mode decomposition

Abstract Radiation‐induced photocurrent in semiconductor devices can be simulated using complex physics‐based models, which are accurate, but computationally expensive. This presents a challenge for implementing device characteristics in high‐level circuit simulations where it is computationally infeasible to evaluate detailed models for multiple individual circuit elements. In this work we demonstrate a procedure for learning compact delayed photocurrent models that are efficient enough to implement in large‐scale circuit simulations, but remain faithful to the underlying physics. Our approach utilizes dynamic mode decomposition (DMD), a system identification technique for learning reduced‐order discrete‐time dynamical systems from time series data based on singular value decomposition. To obtain physics‐aware device models, we simulate the excess carrier density induced by radiation pulses by solving numerically the ambipolar diffusion equation, then use the simulated internal state as training data for the DMD algorithm. Our results show that the significantly reduced‐order delayed photocurrent models obtained via this method accurately approximate the dynamics of the internal excess carrier density—which can be used to calculate the induced current at the device boundaries—while remaining compact enough to incorporate into larger circuit simulations.

Hanson, Joshua↗

NASA NTRS1995

Dynamic neural networks based on-line identification and control of high performance motor drives

In the automated and high-tech industries of the future, there wil be a need for high performance motor drives both in the low-power range and in the high-power range. To meet very straight demands of tracking and regulation in the two quadrants of operation, advanced control technologies are of a considerable interest and need to be developed. In response a dynamics learning control architecture is developed with simultaneous on-line identification and control. the feature of the proposed approach, to efficiently combine the dual task of system identification (learning) and adaptive control of nonlinear motor drives into a single operation is presented. This approach, therefore, not only adapts to uncertainties of the dynamic parameters of the motor drives but also learns about their inherent nonlinearities. In fact, most of the neural networks based adaptive control approaches in use have an identification phase entirely separate from the control phase. Because these approaches separate the identification and control modes, it is not possible to cope with dynamic changes in a controlled process. Extensive simulation studies have been conducted and good performance was observed. The robustness characteristics of neuro-controllers to perform efficiently in a noisy environment is also demonstrated. With this initial success, the principal investigator believes that the proposed approach with the suggested neural structure can be used successfully for the control of high performance motor drives. Two identification and control topologies based on the model reference adaptive control technique are used in this present analysis. No prior knowledge of load dynamics is assumed in either topology while the second topology also assumes no knowledge of the motor parameters.

Rubaai, Ahmed↗

DOE OSTI2024

Multi deep learning-based stochastic microstructure reconstruction and high-fidelity micromechanics simulation of time-dependent ceramic matrix composite response

A multi deep learning-based framework is developed for efficient, automated microstructure reconstruction and generation of stochastic representative volume elements (SRVEs) with periodic boundary conditions (PBCs) for accurate modeling of ceramic matrix composite (CMC) response. The methodology comprises a convolutional neural network coupled with regression layers to act as a vanilla regression network for semantic segmentation of the microstructure, allowing accurate characterization of the phases and their distributions at the microscale. Scanning electron microscope and confocal microscope are used to obtain C/SiNC and SiC/SiNC CMCs micrographs for vanilla regression testing. Microstructure variability in terms of fiber volume fraction and porosity are quantified through the output regression layer, ensuring accurate representation of material variability in SRVE construction. Generative adversarial network (GAN) and its variants are designed to produce high-fidelity SRVE, spanning CMCs microstructure variability space. A circular padding algorithm is developed to generate SRVEs with PBCs during training of GANs. The accuracy of the generated SRVEs is established through micromechanics simulations, where an efficient formulation of the high-fidelity generalized methods of cells (HFGMC) approach is used to compute the effective mechanical properties. Furthermore, an iterative algorithm is implemented in the HFGMC solver to simulate time-dependent deformation of SiC/SiNC subjected to creep loading conditions.

36 MATERIALS SCIENCE↗

DOE OSTI2021

Crystallization of the P 3 Sn 4 Phase upon Cooling P 2 Sn 5 Liquid by Molecular Dynamics Simulation Using a Machine Learning Interatomic Potential

We performed molecular dynamics simulations to study the crystallization of the P3Sn4 phase from P 2 Sn 5 liquid using a machine learning (ML) interatomic potential with desirable efficiency and accuracy. Our results capture the liquid properties of P 2 Sn 5 at 1300 K, which is well above the melting temperature. The phase separation and crystallization are observed when P 2 Sn 5 liquid is cooled down below 832 and 505 K, respectively. The simulation results are in good agreement with the experimentally observed phase transformation behaviors and provide useful insights into the complex nucleation and crystallization process at the details of atomistic scale. Our work also demonstrated that ML interatomic potentials based on neural network deep learning are robust and capable of accurately describing the energetics and kinetics of complex materials through molecular dynamics simulations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

NASA NTRS2012

LEGION: Lightweight Expandable Group of Independently Operating Nodes

LEGION is a lightweight C-language software library that enables distributed asynchronous data processing with a loosely coupled set of compute nodes. Loosely coupled means that a node can offer itself in service to a larger task at any time and can withdraw itself from service at any time, provided it is not actively engaged in an assignment. The main program, i.e., the one attempting to solve the larger task, does not need to know up front which nodes will be available, how many nodes will be available, or at what times the nodes will be available, which is normally the case in a "volunteer computing" framework. The LEGION software accomplishes its goals by providing message-based, inter-process communication similar to MPI (message passing interface), but without the tight coupling requirements. The software is lightweight and easy to install as it is written in standard C with no exotic library dependencies. LEGION has been demonstrated in a challenging planetary science application in which a machine learning system is used in closed-loop fashion to efficiently explore the input parameter space of a complex numerical simulation. The machine learning system decides which jobs to run through the simulator; then, through LEGION calls, the system farms those jobs out to a collection of compute nodes, retrieves the job results as they become available, and updates a predictive model of how the simulator maps inputs to outputs. The machine learning system decides which new set of jobs would be most informative to run given the results so far; this basic loop is repeated until sufficient insight into the physical system modeled by the simulator is obtained.

Burl, Michael C.↗

DOE OSTI2020

Efficient Generalized Boundary Detection Using a Sliding Information Distance

In this work, we present a general machine learning algorithm for boundary detection within general signals based on an efficient, accurate, and robust approximation of the universal normalized information distance. Our approach uses an adaptive sliding information distance (SLID) combined with a wavelet-based approach for peak identification to locate the boundaries. Special emphasis is placed on developing an adaptive formulation of SLID to handle general signals with multiple unknown and/or drifting section lengths. Although specialized algorithms may outperform SLID when domain knowledge is available, these algorithms are limited to specific applications and do not generalize. SLID excels in these cases. We demonstrate the versatility and efficacy of SLID on a variety of signal types, including synthetically generated sequences of tokens, binary executables for reverse engineering applications, and time series of seismic events.

42 ENGINEERING↗

DOE OSTI2023

Machine-learning-based spectral methods for partial differential equations

Spectral methods are an important part of scientific computing’s arsenal for solving partial differential equations (PDEs). However, their applicability and effectiveness depend crucially on the choice of basis functions used to expand the solution of a PDE. The last decade has seen the emergence of deep learning as a strong contender in providing efficient representations of complex functions. In the current work, we present an approach for combining deep neural networks with spectral methods to solve PDEs. In particular, we use a deep learning technique known as the Deep Operator Network (DeepONet) to identify candidate functions on which to expand the solution of PDEs. We have devised an approach that uses the candidate functions provided by the DeepONet as a starting point to construct a set of functions that have the following properties: (1) they constitute a basis, (2) they are orthonormal, and (3) they are hierarchical, i.e., akin to Fourier series or orthogonal polynomials. We have exploited the favorable properties of our custom-made basis functions to both study their approximation capability and use them to expand the solution of linear and nonlinear time-dependent PDEs. The proposed approach advances the state of the art and versatility of spectral methods and, more generally, promotes the synergy between traditional scientific computing and machine learning.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Network Anomaly Detection Using Federated Learning

The internet is turning out to be an integral part of every-one's lives as more and more devices are being connected to serve societal needs. Our work is motivated by two ma-jor observations. Firstly, one drawback of connecting to the network is the threat of network attacks that can compromise users' private information, leading to data loss and adversely affecting productivity. There are several traditional security mechanisms to defend against these attacks, such as firewalls, virtual private networks (VPNs), demilitarized zones (DMZs), and vulnerability scanners. One way to prevent these attacks is early detection and prevention. However, these kinds of architecture do not scale very well because of their centralized nature. Secondly, we observe from heuristics and data set distributions that the majority of the requests made to a server are innocuous. Therefore, almost all server request data sets are highly imbalanced, weighted highly towards the harmless requests.

Marfo, William↗

DOE OSTI2026

Process‐Oriented Calibration of a Turbulence Scheme in the DOE's Global Storm‐Resolving Model Using Machine Learning

A process‐oriented calibration framework is developed for the Simplified Higher‐Order Closure (SHOC) turbulence scheme in DOE's Simple Cloud Resolving E3SM Atmospheric Model (SCREAM). This framework leverages machine learning surrogates and observational constraints to efficiently calibrate SHOC adjustable parameters across two convective regimes: clear‐sky dry convective boundary layer and fair‐weather shallow cumulus clouds from ARM observations. We use perturbed‐parameter ensembles of a doubly periodic version of SCREAM to train surrogates and apply Markov Chain Monte Carlo sampling guided by cost functions based on benchmarking large‐eddy simulations and observations to identify optimized parameter sets that perform well in both regimes. The calibrated SHOC parameters substantially improve boundary‐layer turbulence and cloud boundaries, and modeled cloud fraction and radiative effects align better with observations than the default. These results demonstrate that combining multiple process‐specific convective regimes with machine‐learning surrogates can reduce parametric uncertainties and yield a model more faithful to cloud–turbulence interactions.

58 GEOSCIENCES↗

DOE OSTI2024

Composable optimization and control toolkit for scientific applications

Applications of Artificial Intelligence (AI) and Machine Learning (ML) can improve the computational efficiency and scientific research output. In order to improve interoperability and reuse of AI/ML software, a composable approach is required. This talk presents a composable approach for scientific workflow development that allows seamless integration of various modules developed by independent researchers. These practices will reduce redundant software development by allowing re-use of workflow modules across projects, teams, departments and facilities. We will present three use cases that follow the composable approach namely, Scientific Optimization and Control Toolkit (SOCT), SciDAC QuantOm workflow, and JLab Nuclear Physics experimental workflows. This talk will dive deeper into SOCT and present the details of the composable code development for optimization and control algorithms using reinforcement learning.

Rajput, Kishansingh↗

DOE OSTI2021

An Overview of the Usefulness of Machine Learning Techniques on Network Packet Data

Understanding the health and behavior of a computer network allows for better network efficiency and security. We present an overview of various machine learning techniques for classifying network packet data via packet metadata. While some classical machine learning approaches achieve reasonable results, the most accurate classification can be achieved with deep learning. On the four data sets studied herein, a basic deep learning model achieved at or near 100\% classification accuracy. We also propose a method for determining variable importance as a means for potential transfer learning applications to classifying yet unseen network packet data.

97 MATHEMATICS AND COMPUTING↗

NASA NTRSDate not supplied

Clifford Neural Operators on Atmospheric Data Influenced Partial Differential Equations

Mathematical representations of the atmosphere are key to forecasting and research tasks across Earth science. Numerically solving the underlying partial differential equations(PDEs) of the atmosphere, however, can be difficult and computationally expensive with numerous trade-offs between computing efficiency and accuracy. Utilizing neural net-works to learn approximations of the PDE solutions from the data can help us model complex phenomena more efficiently than traditional numerical schemes. Here, we have applied Clifford algebra-based neural operators for predicting atmospheric variables. Clifford Fourier neural operators are used with two different backbone architectures, ResNet and UNet, on custom data of U10, V10 and surface pressure as well as U500, V500 and Z500. Clifford Fourier neural operators, coupled with ResNet and UNet architectures, are applied to a key reanalysis dataset. Model performance is initially strong, but we observe increasing errors, resulting in the model becoming highly unstable.

Sujit Roy↗

DOE OSTI2025

RingX: Scalable Parallel Attention for Long-Context Learning on HPC

The attention mechanism has become foundational for remarkable AI breakthroughs since the introduction of the Transformer, driving the demand for increasingly longer context to power frontier models such as large-scale reasoning language models and high-resolution image/video generators. However, its quadratic computational and memory complexities present substantial challenges. Current state-of-the-art parallel attention methods, such as ring attention, are widely adopted for long-context training but utilize a point-to-point communication strategy that fails to fully exploit the capabilities of modern HPC network architectures. In this work, we propose ringX, a scalable family of parallel attention methods optimized explicitly for HPC systems. By enhancing workload partitioning, refining communication patterns, and improving load balancing, ringX achieves up to 3.4 × speedup compared to conventional ring attention on the Frontier supercomputer. Optimized for both bi-directional and causal attention mechanisms, ringX demonstrates its effectiveness through training benchmarks of a Vision Transformer (ViT) on a climate dataset and a Generative Pre-Trained Transformer (GPT) model, Llama3 8B. Our method attains an end-to-end training speedup of approximately 1.5 × in both scenarios. To our knowledge, the achieved 38% model FLOPs utilization (MFU) for training Llama3 8B with a 1M-token sequence length on 4,096 GPUs represents one of the highest training efficiencies reported for long-context learning on HPC systems. Our code implementation is available at https://github.com/jqyin/ringX-attention.

Yin, Junqi [ORNL] (ORCID:0000000338435520)↗