Search NASA⌕ Search

SEARCH · Search NASA

Results for “Recommender system”

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 127 records · Page 7

Incremental nonnegative matrix factorization based on correlation and graph regularization for matrix completion

Matrix factorization is widely used in recommendation systems, text mining, face recognition and computer vision. As one of the most popular methods, nonnegative matrix factorization and its incremental variants have attracted much attention. The existing incremental algorithms are established based on the assumption of samples are independent and only update the new latent variable of weighting coefficient matrix when the new sample comes, which may lead to inferior solutions. To address this issue, we investigate a novel incremental nonnegative matrix factorization algorithm based on correlation and graph regularizer (ICGNMF). The correlation is mainly used for finding out those correlated rows to be updated, that is, we assume that samples are dependent on each other. We derive the updating rules for ICGNMF by considering the correlation. We also present tests on widely used image datasets, and show ICGNMF reduces the error by comparing other methods.

Zhang, Xiaoxia↗

Machine learning at the edge to improve in-field safeguards inspections

Artificial intelligence (AI) and machine learning (ML) are near-ubiquitous in day-to-day life; from cars with automated driver-assistance, recommender systems, generative content platforms, and large language chatbots. Implementing AI as a tool for international safeguards could significantly decrease the burden on safeguards inspectors and nuclear facility operators. The use of AI would allow inspectors to complete their in-field activities quicker, while identifying patterns and anomalies and freeing inspectors to focus on the uniquely human component of inspections. Sandia National Laboratories has spent the past two and a half years developing on-device machine learning to develop both a digital and robotic assistant. This combined platform, which we term inspecta, has numerous on-device machine learning capabilities that have been demonstrated at the laboratory scale. Here this work describes early successes implementing AI/ML capabilities to reduce the burden of tedious inspector tasks such as seal examination, information recall, note taking, and more.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗

Impact of water activity on the stability of ettringite

The stability of ettringite as function of temperature and water vapour pressure was studied on synthetic samples using X-ray diffraction, thermogravimetric analysis, sorption balance measurements and different calorimetry techniques. Crystalline ettringite has a water content per mol of Al{sub 2}O{sub 3}, n, varying from 30 to 32 H{sub 2}O, while the amorphous product formed on drying it, metaettringite, has only 9 to 13 H{sub 2}O. Very large hysteresis was observed during the desorption and absorption cycles. The total enthalpy values associated with the decomposition and reformation of ettringite were consistent and similar. During rehydration, different absorption stages were identified and quantified with sorption calorimetry. Due to the hysteresis it was not possible to model the absorption/desorption behaviour using the enthalpies determined by calorimetry, so the values calculated with the van't Hoff equation were used. The area of performance of ettringite-based systems recommended in this work has been extended compared to previously reported data.

36 MATERIALS SCIENCE↗

Greece's reformed EIA system: Evaluating its implementation and potential

Highlights: • Greek nationwide multidimensional EIA System effectiveness evaluation study • Innovation of reform: a central Electronic Environmental Registry (e-ER) • e-ER manages and publish results of the EIA, EMS and Inspection processes. • There is a chasm between new legal provisions and implementation in practice. • Results indicate persistent poor-quality EIA, bureaucracy, lack of transparency. - Abstract: The Memorandum of Understanding (MoU) signed by the Greek government and the troika in 2010 contained numerous austerity policies and requirements for administrative reforms, among which is that of the national Environmental Impact Assessment (EIA) system. Complying with this requirement, Law4014/2011 was passed, aiming at reducing bureaucracy and increasing the country's appeal to investors whilst ensuring environmental protection. The present paper provides the first systematic, multi-scale and multidimensional evaluation of the effectiveness of the new EIA system. The evaluation is based firstly on the results of a Pan-Hellenic survey, involving key EIA consultants and authorities; secondly, on additional, in-depth interviews with EIA specialists, National- Ministry and Regional EIA Authorities; and thirdly, on the content review of 105 Environmental Impact statements. The results highlight the persistence of serious problems despite the reform of the Hellenic EIA System. The data reveal a significant discrepancy between what is set out in law and the actual practice. Although Law 4014/2011 contains progressive and innovative elements (such as adopting a life cycle approach to the EIA process, and mandating the development and use of a central Electronic Environmental Registry (e-ER) to manage and publish results of the EIA process), the recorded failure to implement them in practice results in a non-transparent and ineffective EIA system. Recommendations for improvement are proposed as well as areas for further research.

54 ENVIRONMENTAL SCIENCES↗

TransformerG2G: Adaptive time-stepping for learning temporal graph embeddings using transformers

Dynamic graph embedding has emerged as a very effective technique for addressing diverse temporal graph analytic tasks (i.e., link prediction, node classification, recommender systems, anomaly detection, and graph generation) in various applications. Such temporal graphs exhibit heterogeneous transient dynamics, varying time intervals, and highly evolving node features throughout their evolution. Hence, incorporating long-range dependencies from the historical graph context plays a crucial role in accurately learning their temporal dynamics. In this paper, we develop a graph embedding model with uncertainty quantification, TransformerG2G, by exploiting the advanced transformer encoder to first learn intermediate node representations from its current state (t) and previous context (over timestamps [t–1,t–l], l is the length of context). Moreover, we employ two projection layers to generate lower-dimensional multivariate Gaussian distributions as each node's latent embedding at timestamp t. We consider diverse benchmarks with varying levels of "novelty" as measured by the TEA (Temporal Edge Appearance) plots. Here, our experiments demonstrate that the proposed TransformerG2G model outperforms conventional multi-step methods and our prior work (DynG2G) in terms of both link prediction accuracy and computational efficiency, especially for high degree of novelty. Furthermore, the learned time-dependent attention weights across multiple graph snapshots reveal the development of an automatic adaptive time stepping enabled by the transformer. Importantly, by examining the attention weights, we can uncover temporal dependencies, identify influential elements, and gain insights into the complex interactions within the graph structure. For example, we identified a strong correlation between attention weights and node degree at the various stages of the graph topology evolution.

97 MATHEMATICS AND COMPUTING↗

A deep learning-based battery sizing optimization tool for hybridizing generation plants

Hybrid generation and energy storage systems offer the ability to increase flexibility of the combined asset. This flexibility can be used to increase provision of services already provided by the generation asset, such as timing sale of electricity to the energy market during high price periods, and also enable provision of additional services, such as ancillary services or contribute to resource adequacy. From a generation asset owner perspective, the decision to hybridize includes selecting an energy storage system that, among other factors, maximizes financial performance of the energy storage investment. Yet, existing tools to optimize energy storage sizing are either too rudimentary (i.e., based on “rules of thumb”) or too complex to implement (i.e., require specialized engineering and software knowledge and a high-performance computer to run). This work presents a novel deep learning-based battery sizing optimization tool that is designed to help generation asset owners easily assess preliminary sizing considerations for potential battery investments to hybridize their generation facility. The tool uses deep learning to predict revenue over a broad search space of potential battery sizes, estimates capital and operating costs (including accounting for battery degradation), and computes financial performance of each potential battery system investment, recommending a system with maximum financial performance. The tool is tested and validated for hydropower assets. Finally, this tool will help a greater cross-section of industry consider investments in battery systems, increasing their revenue and helping them compete in rapidly evolving electrify markets.

13 HYDRO ENERGY↗

nmRanalysis: An Open-Source Web Application for Semi-automated NMR Metabolite Profiling

Though data acquisition and initial signal pre-processing of nuclear magnetic resonance (NMR) spectra have achieved high degrees of automation, downstream processing - specifically the profiling of spectra - has bottlenecked the overall NMR analysis workflow. Several efforts have been made to mitigate this bottleneck, but these solutions often trade an increase in automation for limitations elsewhere. Here, in this technical note, we introduce nmRanalysis, a user-friendly web-application that integrates the strengths of existing profiling tools for a more automated profiling workflow. nmRa-nalysis additionally incorporates novel features, including a machine-learning-driven recommender system for me-tabolite identification, further increasing the utility of nmRanalysis over the individual tools that it incorporates.

Flores, Javier E. [Pacific Northwest National Labo↗

Anomaly detection in the Zwicky Transient Facility DR3

We present results from applying the SNAD anomaly detection pipeline to the third public data release of the Zwicky Transient Facility (ZTF DR3). The pipeline is composed of three stages: feature extraction, search of outliers with machine learning algorithms, and anomaly identification with followup by human experts. Our analysis concentrates in three ZTF fields, comprising more than 2.25 million objects. A set of four automatic learning algorithms was used to identify 277 outliers, which were subsequently scrutinized by an expert. From these, 188 (68 per cent) were found to be bogus light curves – including effects from the image subtraction pipeline as well as overlapping between a star and a known asteroid, 66 (24 per cent) were previously reported sources whereas 23 (8 per cent) correspond to non-catalogued objects, with the two latter cases of potential scientific interest (e.g. one spectroscopically confirmed RS Canum Venaticorum star, four supernovae candidates, one red dwarf flare). Moreover, using results from the expert analysis, we were able to identify a simple bi-dimensional relation that can be used to aid filtering potentially bogus light curves in future studies. We provide a complete list of objects with potential scientific application so they can be further scrutinised by the community. These results confirm the importance of combining automatic machine learning algorithms with domain knowledge in the construction of recommendation systems for astronomy. Our code is publicly available.

79 ASTRONOMY AND ASTROPHYSICS↗

Challenges and Opportunities in Deep Reinforcement Learning With Graph Neural Networks: A Comprehensive Review of Algorithms and Applications

Deep reinforcement learning (DRL) has empowered a variety of artificial intelligence fields, including pattern recognition, robotics, recommendation-systems, and gaming. Similarly, graph neural networks (GNN) have also demonstrated their superior performance in supervised learning for graph-structured data. In recent times, the fusion of GNN with DRL for graph-structured environments has attracted a lot of attention. Here, this paper provides a comprehensive review of these hybrid works. These works can be classified into two categories: (1) algorithmic enhancement, where DRL and GNN complement each other for better utility; (2) application-specific enhancement, where DRL and GNN support each other. This fusion effectively addresses various complex problems in engineering and life sciences. Based on the review, we further analyze the applicability and benefits of fusing these two domains, especially in terms of increasing generalizability and reducing computational complexity. Finally, the key challenges in integrating DRL and GNN, and potential future research directions are highlighted, which will be of interest to the broader machine learning community.

97 MATHEMATICS AND COMPUTING↗

Novel Proposals for FAIR, Automated, Recommendable, and Robust Workflows

Lightning talks of the Workflows in Support of Large-Scale Science (WORKS) workshop are a venue where the workflow community (researchers, developers, and users) can discuss work in progress, emerging technologies and frameworks, and training and education materials. This paper summarizes the WORKS 2022 lightning talks, which cover five broad topics: data integrity of scientific workflows; a machine learning-based recommendation system; a Python toolkit for running dynamic ensembles of simulations; a cross-platform, high-performance computing utility for processing shell commands; and a meta(data) framework for reproducing hybrid workflows.

Abhinit, Ishan↗

Towards Compact Neural Networks via End-to-End Training: A Bayesian Tensor Approach with Automatic Rank Determination

Post-training model compression can reduce the inference costs of deep neural networks, but uncompressed training still consumes enormous hardware resources and energy. To enable low-energy training on edge devices, it is highly desirable to directly train a compact neural network from scratch with a low memory cost. Low-rank tensor decomposition is an effective approach to reduce the memory and computing costs of large neural networks. However, directly training low-rank tensorized neural networks is a very challenging task because it is hard to determine a proper tensor rank a priori, and the tensor rank controls both model complexity and accuracy. Here, this paper presents a novel end-to-end framework for low-rank tensorized training. We first develop a Bayesian model that supports various low-rank tensor formats (e.g., CANDECOMP/PARAFAC, Tucker, tensor-train, and tensor-train matrix) and reduces neural network parameters with automatic rank determination during training. Then we develop a customized Bayesian solver to train large-scale tensorized neural networks. Our training methods shows orders-of-magnitude parameter reduction and little accuracy loss (or even better accuracy) in the experiments. On a very large deep learning recommendation system with over 4.2 ×10 9 model parameters, our method can reduce the parameter number to 1.6 ×10 5 automatically in the training process (i.e., by 2.6 ×10 4 times) while achieving almost the same accuracy. Code is available at https://github.com/colehawkins/bayesian-tensor-rank-determination.

compact neural networks↗

Hybrid-BPR (Bayesian Personalized Ranking with Feature Embeddings and Explicit Negative Sampling) [SWR-26-039]

Hybrid-BPR is a Python library for Bayesian Personalized Ranking (BPR) with two key capabilities that go beyond standard BPR implementations: 1. User and item feature embeddings - incorporate content-based signals (genres, tags, metadata) alongside collaborative filtering. 2. Implicit negative interactions - use observed non-interactions (e.g. viewed-but-not-clicked) as negative training signal instead of random sampling from the full item space. The software is built for recommender systems research with MLflow experiment tracking, parallel hyperparameter sweeps, and standard ranking metrics.

Sandhu, Rimple [National Laboratory of the Rockies↗

Design considerations for a Digital Twin built to improve nitrification performance at a water resource recovery facility

A Digital Twin built around Activated Sludge Model No. 1 was deployed at a full-scale water resource recovery facility. Its design included a waste rate recommender system based on automatic scenario analyses, where influent loads and waste rates are varied to determine their impact on nitrification. At the same frequency as these scenario analyses, scheduled auto-calibrations allow for nitrifier maximum specific growth rate (μmax-NITO) soft sensing, the only kinetic parameter shown to require adjustment if the objective is aeration tank effluent ammonia forecasting accuracy. By integrating temperature forecasting over the next three sludge ages, this Digital Twin approach creates opportunities for advancing waste rate decisions in anticipation of seasonal temperature changes, optimizing ammonia control authority under varying influent loads, and furnishing valuable insights for future capital projects requiring nitrifier kinetic understanding and modelling.

54 ENVIRONMENTAL SCIENCES↗

Roadmap for thermal property measurements of Molten Salt Reactor systems

Thermal properties of molten salts, both with and without fissile and fertile material, are fundamental to modeling and simulation that supports design, construction, operation, and accident progression evaluation of molten salt reactors. This roadmap provides guidance to the research community on the types and amounts of required data for base, or carrier, salt systems, recommends simulated burnup fuel salt compositions to evaluate, and provides a technical foundation for a focused salt property evaluation matrix. The property values generated from both computation and experimental methods are used to develop models that constitute the Molten Salt Thermal Properties Database (MSTDB). While the building blocks of MSTDB are fundamental in nature, its data facilitates constructing engineering models that allow stakeholders to investigate the behavior of specific compositions of interest and supports broader modeling and simulation of MSRs through coupling to multi-physics, multi-scale mass accountancy tools.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Orbit-spectrum sharing between the fixed-satellite and broadcasting-satellite services with applications to 12 GHz domestic systems

A systematic, tutorial analysis of the general problem of orbit-spectrum sharing among inhomogeneous satellite system is presented. Emphasis is placed on extrapolating and applying the available data on rain attenuation and on reconciling differences in the results of various measurements of the subjective effects of interference on television picture quality. An analytic method is presented for determining the approximate values of the intersatellite spacings required to keep mutual interference levels within prescribed limits when many dissimilar satellites share the orbit. A computer model was developed for assessing the interference compatibility of arbitrary configurations of large numbers of geostationary satellite systems. It is concluded that the band from 11.7 c GHz can be shared effectively by broadcasting-satellite and fixed-satellite systems. Recommendations for future study are included.

Reinhart, E. E.↗

Earth resources experiment package sensor performance evaluation. Volume 2: S191

An S191 performance summary based on ground testing and orbital operations in terms of pertinent parameters is provided. Additional tasks covering S191 short-wavelength radiometric calibration, data-acquisition camera, and viewfinder tracking system design criteria, radiometric comparison of S190A, S191, and S192, and wavelength calibration data are included as supplemental analyses. Descriptions of techniques employed in the performance analyses are also discussed. Final results of the evaluations in terms of significant performance degradation, sensor and system anomalies, and achieved performance are presented. Conclusions were based on S191 performance and interaction of S191 with the EREP system. Recommendations for additional analyses and improvements in design and operation are presented.

Kenney, G. P.↗

Design of a video system providing optimal visual information for controlling payload and experiment operations with television

A program was conducted which included the design of a set of simplified simulation tasks, design of apparatus and breadboard TV equipment for task performance, and the implementation of a number of simulation tests. Performance measurements were made under controlled conditions and the results analyzed to permit evaluation of the relative merits (effectivity) of various TV systems. Burden factors were subsequently generated for each TV system to permit tradeoff evaluation of system characteristics against performance. For the general remote operation mission, the 2-view system is recommended. This system is characterized and the corresponding equipment specifications were generated.

Source record↗

Tug and payload-to-Orbiter interface requirements

This paper presents results of a study to identify, evaluate, and develop Tug plus payload-to-Orbiter accommodation requirements. Interface areas include structural, mechanical, fluids, environmental, avionics, and safety. Each interface area was investigated to determine the best physical and operational method of accomplishing the required functions with an overriding goal of establishing sample and flexible interface requirements suitable for Tug, Tug payloads, and other cargo. The recommended system for supporting the Tug and its payload from the Orbiter employs a cylindrical load-carrying structure called a deployment adapter that has a rotational capability for Tug deployment. The design and operation of the deployment adapter and other peripheral interface equipment such as crew compartment monitor and control panels and cargo bay fluid and electrical interface panels are discussed.

Bock, E. H.↗