Search NASA⌕ Search

SEARCH · Search NASA

Results for “transfer 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 271 records · Page 15

PemNet: A Transfer Learning-Based Modeling Approach of High-Temperature Polymer Electrolyte Membrane Electrochemical Systems

Widespread adoption of high-temperature electrochemical systems such as polymer electrolyte membrane fuel cells (HT-PEMFCs) requires models and computational tools for accurate optimization and guiding new materials for enhancing fuel cell performance and durability. Furthermore, while robust and better suited for extrapolation, knowledge-based modeling has limitations as it is time-consuming and requires information about the system that is not always available (e.g., material properties and interfacial behavior between different materials). Data-driven modeling, on the other hand, is easier to implement but often necessitates large datasets that could be difficult to obtain. In this contribution, knowledge-based modeling and data-driven modeling are combined by implementing a few-shot learning (FSL) approach. A knowledge-based model originally developed for a HT-PEMFCs was used to generate simulated data (887,735 points) and used to pretrain a neural network source model tuned via a genetic algorithm-based AutoML. Then, experimental datasets from HT-PEMFCs with different materials and operating conditions (~50 points each) were used to train six target models via FSL. Models for the unseen data reached high accuracies in all cases (rRMSE < 10%).

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Reweighting configurations generated by transferable, machine learned models for protein sidechain backmapping

Multiscale modeling requires the linking of models at different levels of detail, with the goal of gaining accelerations from lower fidelity models while recovering fine details from higher resolution models. Communication across resolutions is particularly important in modeling soft matter, where tight couplings exist between molecular-level details and mesoscale structures. While multiscale modeling of biomolecules has become a critical component in exploring their structure and self-assembly, backmapping from coarse-grained to fine-grained, or atomistic, representations presents a challenge, despite recent advances through machine learning. A major hurdle, especially for strategies utilizing machine learning, is that backmappings can only approximately recover the atomistic ensemble of interest. We demonstrate conditions for which backmapped configurations may be reweighted to exactly recover the desired atomistic ensemble. By training separate decoding models for each sidechain type, we develop an algorithm based on normalizing flows and geometric algebra attention to autoregressively propose backmapped configurations for any protein sequence. Critical for reweighting with modern protein force fields, our trained models include all hydrogen atoms in the backmapping and make probabilities associated with atomistic configurations directly accessible. We also demonstrate, however, that reweighting is extremely challenging despite state-of-the-art performance on recently developed metrics and generation of configurations with low energies in atomistic protein force fields. Through detailed analysis of configurational weights, we show that machine-learned backmappings must not only generate configurations with reasonable energies, but also correctly assign relative probabilities under the generative model. These are broadly important considerations in generative modeling of atomistic molecular configurations.

Monroe, Jacob I. [Univ. of Arkansas, Fayetteville,↗

SMALE: Enhancing Scalability of Machine Learning Algorithms on Extreme-Scale Computing Platforms

Deployment and execution of machine learning tasks on extreme-scale computing platforms face several significant technical challenges: 1) High computing cost incurred by dense networks – The computing workload of deep networks with densely-connected topology increases rapidly with the network size, imposing a non-scalable computing model of extreme-scale computing platforms; 2) Non-optimized workload distribution – Many advanced deep learning algorithms, e.g., sparsification and irregular net-work topology, produce very unbalanced workload distribution on extreme-scale computing platforms. The computation efficiency is greatly hindered by the incurred data and computation redundancies as well as long tails of the node with extensive workload; 3) Constraints in data movement and I/O bottle-neck – Inter-node data movement in extreme-scale computing platforms are associated with high energy and latency costs, and subject to the constraints of I/O bandwidth; and 4) Generalization of algorithm realization and acceleration on computing platforms – The large varieties of machine learning algorithms and structures of extreme-scale computing platforms make the derivation of a generalized algorithm realization and acceleration method very challenging, which, however, is the requirement by domain scientists and interested users. We call the above challenges Smale’s Problems in Machine Learning and Understanding for High-Performance Computing Scientific Discovery. The objective of our three-year research project is to develop a holistic innovation set at structure, assembly, and acceleration layers of machine learning algorithms to address the above challenges in algorithm deployment and execution. Three tasks are particularly performed, including: At the algorithm structure level, we investigate the techniques that can structurally sparsify on the topology of deep networks for computing workload reduction. We also study clustering and pruning techniques that can optimize the workload distributions over the extreme-scale computing platforms; At the algorithm assembly level, we derive a unified learning framework for unsupervised transfer learning and dynamic growing capabilities. Novel training methods are also exploited to enhance the training efficiency of the proposed framework; At the algorithm acceleration level, we will develop a series of techniques that can accelerate the computation of sparse matrix operations, which are one of the core executions in deep learning and optimize memory access of the concerned platforms. Our proposed techniques attack the fundamental problems in machine learning algorithms running on extreme-scale computing platforms by vertically integrating the solutions at three closely entangled layers, paving the long-term scaling path of machine learning applications under DOE context. Three tasks corresponding to the above respective research orientations are performed during the three-year project period with our collaborators at ORNL. The outcome of the proposed project is anticipated to form a holistic solution set of novel algorithms and network topologies, efficient training techniques, and fast acceleration methods to promote the computing scalability of the machine learning applications of particular interest to DOE.

97 MATHEMATICS AND COMPUTING↗

The SRNL technical assistance program: transferring lessons learned from two decades of remedial successes to the DOE complex - 15438

The Savannah River Site has an established history of developing implementable risk reduction strategies, developing and deploying innovative technological solutions, and collaboratively achieving realistic and protective environmental end states for complex environmental problems. Part of this success results from the strong historical collaboration between site environmental restoration groups and the Savannah River National Laboratory which results in the promulgation of innovative technical approaches and technologies from the laboratory to ongoing remediation campaigns. Much of the underlying research and development effort was funded by DOE's Office on Environmental Management through a series of applied technology demonstration projects (Integrated Demonstration Project, Integrated Programs and the Applied Field Research Site). Concomitant with the research efforts, SRNL has managed a dynamic and efficient national program that provides teams of technical experts with a broad experience base to recommend strategies that reduce risk and/or technical uncertainty for challenging environmental problems at other DOE facilities. At many of these sites, standard remedial approaches (e.g., excavation, pump and treat, soil vapor extraction) have either failed or proved to be too costly, inadequate, or ineffective. Alternatively, other sites struggle with implementation of innovative but less well understood approaches. The technical assistance teams use a structured process that based on several basic concepts, specifically, development of site specific technical frameworks, and careful matching of remedies to site-specific chemical, geological and physical conditions. Development of technical frameworks is a key strategy to apply basic science to an applied field problem. When directed toward understanding complex real-world environmental remediation challenges, frameworks are tools that support practical identification and incorporation of the key-controlling scientific processes and principles. Frameworks can also be used to minimize technical risks, encourage efficiency and effectiveness, and provide the basis for innovative and creative solutions. Efficient and effective environmental cleanup also requires matching the character of remediation and stabilization methods to the nature of the target zone of contamination as the nature of the target zone evolves through the life of the remedial project. A contaminant plume can be divided into the following zones: the source zone, the impact zone, and the transitional zone that then can be used to identify classes of technologies that are appropriate to that zone. For example, physical and chemical methods (e.g., trapping, immobilization, destruction, or isolation) that directly address the source contaminants are often appropriate for the disturbed zone during the remedial process. A variety of methods that include both active treatments (e.g., pump and treat or active bioremediation) and enhanced attenuation technologies (e.g., geochemical manipulation or reactive barriers) are often suitable for the primary contamination zone or impact zone. Various strategies based on natural attenuation processes may be applicable to the primary contamination zone and these methods are typically applied for the transition or baseline portions of the plume. Careful matching of remedies to site-specific conditions is critical to long-term success in environmental cleanup and restoration. The matching process facilitates selection of technologies with particular strengths that align with real-world needs and constraints, encourages strategic use of multiple or combined technologies to address major plume sub-domains, and supports transitioning technologies in space and time as remediation progresses. Since 2006, the SRNL Technical Assistance program has focused on providing support across the DOE complex. During this time, over 25 teams have visited eleven DOE sites and made recommendations that yielded an estimated cost savings of $100 M. Examples will be provided that illustrate key aspects and significant successes of the technical assistance program. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

New data-driven approach to bridging power system protection gaps with deep learning

Protection is a critical function in power systems to avoid equipment damage, maintain personnel safety, and support system reliability. However, current protective relay technology cannot adequately protect equipment and personnel from effects of some events; these deficiencies are termed protection gaps. In this paper, a data-driven approach is proposed to complement traditional protection technology and distinguish fault conditions from transients caused by normal operations. A combined convolutional neural network and long short-term memory (CNN-LSTM) network is implemented to achieve data translation invariance and capture the temporal correlation of the time-series input data. As a result, the data-driven method can accurately detect system faults despite variation and noise in the input data. In addition, using the CNN-LSTM--based method avoids the complicated, manual feature extraction procedure required by many traditional data-driven methods. The effectiveness of the proposed approach is tested on two kinds of protection gaps: high-impedance faults and transformer inter-turn faults. Lastly, a transfer learning method is also proposed to address the common issue of data-driven methods for which real-world training data are scarce. Extensive study results demonstrate that the proposed approach can accurately bridge power system protection gaps.

42 ENGINEERING↗

Leveraging Multitime Hamilton–Jacobi PDEs for Certain Scientific Machine Learning Problems

Hamilton-Jacobi partial differential equations (HJ PDEs) have deep connections with a wide range of fields, including optimal control, differential games, and imaging sciences. By considering the time variable to be a higher dimensional quantity, HJ PDEs can be extended to the multi-time case. In this paper, we establish a novel theoretical connection between specific optimization problems arising in machine learning and the multi-time Hopf formula, which corresponds to a representation of the solution to certain multi-time HJ PDEs. Through this connection, we increase the interpretability of the training process of certain machine learning applications by showing that when we solve these learning problems, we also solve a multi-time HJ PDE and, by extension, its corresponding optimal control problem. As a first exploration of this connection, we develop the relation between the regularized linear regression problem and the Linear Quadratic Regulator (LQR). We then leverage our theoretical connection to adapt standard LQR solvers (namely, those based on the Riccati ordinary differential equations) to design new training approaches for machine learning. Lastly, we provide some numerical examples that demonstrate the versatility and possible computational advantages of our Riccati-based approach in the context of continual learning, post-training calibration, transfer learning, and sparse dynamics identification.

97 MATHEMATICS AND COMPUTING↗

Technology Transfer

The objective of this summer's work was to attempt to enhance Technology Application Group (TAG) ability to measure the outcomes of its efforts to transfer NASA technology. By reviewing existing literature, by explaining the economic principles involved in evaluating the economic impact of technology transfer, and by investigating the LaRC processes our William & Mary team has been able to lead this important discussion. In reviewing the existing literature, we identified many of the metrics that are currently being used in the area of technology transfer. Learning about the LaRC technology transfer processes and the metrics currently used to track the transfer process enabled us to compare other R&D facilities to LaRC. We discuss and diagram impacts of technology transfer in the short run and the long run. Significantly, it serves as the basis for analysis and provides guidance in thinking about what the measurement objectives ought to be. By focusing on the SBIR Program, valuable information regarding the strengths and weaknesses of this LaRC program are to be gained. A survey was developed to ask probing questions regarding SBIR contractors' experience with the program. Specifically we are interested in finding out whether the SBIR Program is accomplishing its mission, if the SBIR companies are providing the needed innovations specified by NASA and to what extent those innovations have led to commercial success. We also developed a survey to ask COTR's, who are NASA employees acting as technical advisors to the SBIR contractors, the same type of questions, evaluating the successes and problems with the SBIR Program as they see it. This survey was developed to be implemented interactively on computer. It is our hope that the statistical and econometric studies that can be done on the data collected from all of these sources will provide insight regarding the direction to take in developing systematic evaluations of programs like the SBIR Program so that they can reach their maximum effectiveness.

Smith, Nanette R.↗

ABF DFO with Technology Holding, Inc.

This Agile BioFoundry Directed Funding Opportunity project with Technology Holding and partners focuses on the development of both a strain of Pseudomonas putida KT2440 and a corresponding bioprocess to convert cellulosic sugars to beta-ketoadipic acid, which can be used in performance nylons and polyesters. Our approach follows the Design-Build-Test-Learn cycle wherein we have transferred learnings from muconic acid production in P. putida to develop a glucose and xylose-utilizing beta-ketoadipic acid production strain. This strain achieves 65 g/L of beta-ketoadipic acid at 0.7 g/L/hr and a C-mol yield of 0.40. We are currently on-boarding arabinose utilization as well. To identify non-intuitive strain modifications as well, we are deploying a beta-ketoadipic acid biosensor and building randomly barcoded transposon insertion sequencing (RB-TnSeq) libraries and gene over-expression libraries in beta-ketoadipic acid production strains. Moreover, we are using global metabolomics and other systems biology tools to identify off-target pathways. Lastly, we are scaling up beta-ketoadipic acid production to kg-scale production for Technology Holding to evaluate in performance polymers with their partners.

beta-ketoadipic acid↗

Deep reinforcement learning control of hydraulic fracturing

Hydraulic fracturing is a technique to extract oil and gas from shale formations, and obtaining a uniform proppant concentration along the fracture is key to its productivity. Recently, various model predictive control schemes have been proposed to achieve this objective. But such controllers require an accurate and computationally efficient model which is difficult to obtain given the complexity of the process and uncertainties in the rock formation properties. In this article, we design a model-free data-based reinforcement learning controller which learns an optimal control policy through interactions with the process. Deep reinforcement learning (DRL) controller is based on the Deep Deterministic Policy Gradient algorithm that combines Deep-Q-network with actor-critic framework. In addition, we utilize dimensionality reduction and transfer learning to quicken the learning process. We show that the controller learns an optimal policy to obtain uniform proppant concentration despite the complex nature of the process while satisfying various input constraints.

42 ENGINEERING↗

Identifying common stored product insects using automated deep learning methods

Monitoring stored product insect pests is a common practice for post-harvest management of stored grain and grain-based commodities, which helps ensure product quality from harvest to final consumer. Current methods of sampling and monitoring can be time-consuming, labor-intensive, expensive and require expertise in insect identification. Therefore, this study aims to develop an image-based automated identification system for common stored product insect species using deep-learning methods. Top-down images of the common stored product adult insect species of Rhyzopertha dominica, Cryptolestes ferrugineus, Tribolium castaneum, Sitophilus oryzae, and Oryzaephilus surinamensis were acquired and analyzed. Deep learning-based, state-of-the-art Convolutional Neural Networks (CNN) models (ResNet-50, MobileNet-v2, DarkNet-53, and EfficientNet-b0) were fine-tuned with a transfer learning approach to classify the insect species. All models were able to correctly identify the insect species with at least 96% accuracy and with few misclassifications. One issue with trained CNNs is that they do not explain the reasoning for the classification and are often called a “black box”. Therefore, visualization methods called Gradient-weighted Class Activation Mapping (Grad-CAM) were implemented to explore the black box network. The Grad-CAM uses heat maps to highlight the major image features that the network focused on to make insect species predictions. The Grad-CAM verifies the network's prediction and also helps improve network performance. This study contributes to the overall goal of developing a camera-based system for monitoring stored grain insects. As a result, the developed system would empower warehouse, flour mills, and other food facilities with a tool to quickly and accurately identify insect species in stored product environments and could be implemented as part of a close to real-time monitoring system.

60 APPLIED LIFE SCIENCES↗

Watch and learn—a generalized approach for transferrable learning in deep neural networks via physical principles

Transfer learning refers to the use of knowledge gained while solving a machine learning task and applying it to the solution of a closely related problem. Such an approach has enabled scientific breakthroughs in computer vision and natural language processing where the weights learned in state-of-the-art models can be used to initialize models for other tasks which dramatically improve their performance and save computational time. Here we demonstrate an unsupervised learning approach augmented with basic physical principles that achieves fully transferrable learning for problems in statistical physics across different physical regimes. By coupling a sequence model based on a recurrent neural network to an extensive deep neural network, we are able to learn the equilibrium probability distributions and inter-particle interaction models of classical statistical mechanical systems. Our approach, distribution-consistent learning, DCL, is a general strategy that works for a variety of canonical statistical mechanical models (Ising and Potts) as well as disordered interaction potentials. Using data collected from a single set of observation conditions, DCL successfully extrapolates across all temperatures, thermodynamic phases, and can be applied to different length-scales. This constitutes a fully transferrable physics-based learning in a generalizable approach.

97 MATHEMATICS AND COMPUTING↗

BuildingsBench: A Large-Scale Dataset of 900K Buildings and Benchmark for Short-Term Load Forecasting

The BuildingsBench datasets consist of: - Buildings-900K: A large-scale dataset of 900K buildings for pretraining models on the task of short-term load forecasting (STLF). Buildings-900K is statistically representative of the entire U.S. building stock. - 7 real residential and commercial building datasets for benchmarking two downstream tasks evaluating generalization: zero-shot STLF and transfer learning for STLF. Buildings-900K can be used for pretraining models on day-ahead STLF for residential and commercial buildings. The specific gap it fills is the lack of large-scale and diverse time series datasets of sufficient size for studying pretraining and finetuning with scalable machine learning models. Buildings-900K consists of synthetically generated energy consumption time series. It is derived from the NREL End-Use Load Profiles (EULP) dataset (see link to this database in the links further below). However, the EULP was not originally developed for the purpose of STLF. Rather, it was developed to "...help electric utilities, grid operators, manufacturers, government entities, and research organizations make critical decisions about prioritizing research and development, utility resource and distribution system planning, and state and local energy planning and regulation." Similar to the EULP, Buildings-900K is a collection of Parquet files and it follows nearly the same Parquet dataset organization as the EULP. As it only contains a single energy consumption time series per building, it is much smaller (~110 GB). BuildingsBench also provides an evaluation benchmark that is a collection of various open source residential and commercial real building energy consumption datasets. The evaluation datasets, which are provided alongside Buildings-900K below, are collections of CSV files which contain annual energy consumption. The size of the evaluation datasets altogether is less than 1GB, and they are listed out below: 1. ElectricityLoadDiagrams20112014 2. Building Data Genome Project-2 3. Individual household electric power consumption (Sceaux) 4. Borealis 5. SMART 6. IDEAL 7. Low Carbon London A README file providing details about how the data is stored and describing the organization of the datasets can be found within each data lake version under BuildingsBench.

Array↗

Deep-learning-based domain adaptation for cavity fault prediction at Jefferson Laboratory

Superconducting radio-frequency (SRF) cavities are the core components of the Continuous Electron Beam Accelerator Facility (CEBAF) at Jefferson Lab, providing high-power electron beams for nuclear physics experiments. The facility comprises 418 SRF cavities, and any fault in these cavities can lead to interruptions in the electron beam supply. Cavity faults are the leading cause of beam trips in CEBAF. Predicting and mitigating those faults before onset can help maintain normal operation. Existing models face challenges in distinguishing between normal and fault signals when changes occur in the underlying time-series data, from changes in control software, operational parameters, or the environment. This work proposes a deep learning domain adaptation model that leverages transfer learning to address fault prediction challenges by improving accuracy. The model is trained and fine-tuned using a dataset collected for faulty and normal operation using a data acquisition system in CEBAF. Our deep learning-based domain adaptation model achieves a prediction accuracy of 89.61% of the fault and normal signals. The developed model effectively predicts normal running signals compared to the baseline approach without domain adaptation. This capacity is essential for the fault prediction task in the CEBAF because of heavily imbalanced data containing vast amounts of normal signals. The model performs well for predicting faults several hundred milliseconds before the fault onset compared to other models where no adaptation is applied. Incorporating deep learning-based domain adaptation techniques will significantly improve the fault prediction performance.

Rahman, Md Monibor [Old Dominion Univ., Norfolk, V↗

Deffe: A Data-efficient Framework for Performance Characterization in Domain-Specific Computing

Predicting workload performance is a crucial task for many architecture and system research. In this paper we present De￿e, a framework to estimate the workload performances under varying architectural configurations. The infrastructure component of Deffe is based on scalable and easy-to-use open-source software components. By casting the performance modeling as transfer learning tasks, the modeling component of Deffe can leverage the learned knowledge on one workload, and “transfer” it to a new workload. Extensive experimental results show that the method can achieve superior testing accuracy with an effective reduction of 32-80x in terms of the amount of required training data.

Liu, Frank↗