Search NASA⌕ Search

SEARCH · Search NASA

Results for “Collaborative 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 289 records · Page 16
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Towards Trust-Augmented Visual Analytics for Data-Driven Energy Modeling

The promise of data-driven predictive modeling is being increasingly realized in various science and engineering disciplines, where experts are used to the more conventional, simulation-driven modeling practices. However, trust remains a bottleneck for greater adoption of machine learning-based models for domain experts, who might not be necessarily trained in data science. In this paper, we focus on the building energy domain, where physics-based simulations are being complemented or replaced by machine learning-based methods for forecasting energy supply and demand at various spatio-temporal scales. We study the trust problem in close collaboration with energy scientists and engineers and describe how visual analytics can be leveraged for alleviating this trust bottleneck for stakeholders with varying degrees of expertise and analytics goals in this domain.

Kandakatla, Akshith R.↗
DOE OSTI2016

The Livermore Brain: Massive Deep Learning Networks Enabled by High Performance Computing

The proliferation of inexpensive sensor technologies like the ubiquitous digital image sensors has resulted in the collection and sharing of vast amounts of unsorted and unexploited raw data. Companies and governments who are able to collect and make sense of large datasets to help them make better decisions more rapidly will have a competitive advantage in the information era. Machine Learning technologies play a critical role for automating the data understanding process; however, to be maximally effective, useful intermediate representations of the data are required. These representations or “features” are transformations of the raw data into a form where patterns are more easily recognized. Recent breakthroughs in Deep Learning have made it possible to learn these features from large amounts of labeled data. The focus of this project is to develop and extend Deep Learning algorithms for learning features from vast amounts of unlabeled data and to develop the HPC neural network training platform to support the training of massive network models. This LDRD project succeeded in developing new unsupervised feature learning algorithms for images and video and created a scalable neural network training toolkit for HPC. Additionally, this LDRD helped create the world’s largest freely-available image and video dataset supporting open multimedia research and used this dataset for training our deep neural networks. This research helped LLNL capture several work-for-others (WFO) projects, attract new talent, and establish collaborations with leading academic and commercial partners. Finally, this project demonstrated the successful training of the largest unsupervised image neural network using HPC resources and helped establish LLNL leadership at the intersection of Machine Learning and HPC research.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

The role of international collaboration in the United States geologic disposal research program

Abstract. More than a decade ago, the United States disposal program discontinued all research activities focused on the unsaturated fractured tuff formation at Yucca Mountain as the geologic disposal site for spent fuel and high-level radioactive waste. A new research and development (R&D) program was initiated to provide a sound technical basis for alternative disposal options across clay, crystalline, and salt rocks. The goals of this broad program were (and still are) to (1) increase confidence in the robustness of generic disposal concepts, (2) develop the science and engineering tools needed to support disposal concept implementation, and (3) conduct R&D on the direct disposal of existing dual-purpose (storage and transportation) canisters. Recognizing the benefits of international collaboration toward the common goal of safely and efficiently managing the back end of the nuclear fuel cycle, the program emphasized international cooperation as an effective strategy for sharing information and knowledge. In a multi-laboratory effort coordinated by Lawrence Berkeley National Laboratory, the United States Department of Energy (DOE) program established formal and informal cooperation partnerships with several international initiatives and institutions and developed a number of collaborative R&D activities in important research areas, such as engineered barrier integrity, near-field perturbations, radionuclide transport, performance assessment, and methods for characterization and monitoring of engineered and natural barriers. This presentation gives an overview of these R&D activities, with a specific focus on activities that improve our current understanding of the coupled thermal–hydrological–mechanical–chemical (THMC) processes occurring in engineered and natural barriers. We start with a brief review of selected international collaboration initiatives and then describe a few specific collaboration projects. We focus specifically on such studies that use experimental data sets provided by international research cooperation for joint modeling work to increase confidence in performance-relevant predictions of coupled processes. Overall, the focus on international collaboration has allowed deep engagement of US researchers with the international waste management R&D community in terms of best practices, new scientific advances, state-of-the-art simulation tools, new monitoring and performance confirmation approaches, and lessons learned. The joint R&D with international researchers, worldwide sharing of knowledge and experience, and access to relevant data and experiments from a variety of host rocks have helped our researchers to significantly improve their understanding of the current technical basis for disposal in a range of potential host rock environments. International collaboration also provides ample opportunity for training and educating junior staff that are well suited to move the United States disposal research program forward into the next decades, a promising avenue for developing a next-generation workforce of disposal scientists.

Birkholzer, Jens T.↗
DOE OSTI2018

Fit for Purpose Remote Technology Decommissioning Solutions - 18630

The Magnox Decommissioning Project is actively progressing clean-up and decommissioning of 12 reactor sites across England, Scotland and Wales. Funded and directed by the United Kingdom's Nuclear Decommissioning Authority (NDA) and managed by the Cavendish Fluor Partnership (CFP), the Magnox Decommissioning Project introduces a wide range of radiological and industrial hazards warranting deployment of remote technologies to safely de-inventory, decontaminate and decommission a vast array of nuclear facilities. The Magnox Decommissioning Project encompasses 22 former Magnox nuclear power production reactors across 10 reactor sites and five research/test reactors at two former research sites. Decommissioning activities have been underway at some sites for nearly 20 years with early focus on retrieving and stabilising wastes and draining and stabilising spent fuel ponds. The Magnox decommissioning strategy invokes a programme perspective to drive consistency in tackling similar decommissioning and waste challenges across multiple sites. Programme central teams have worked with technology suppliers to foster learning from industry experience to develop and deploy tooling and robotics. The programme approach has also facilitated learning from one site to the next. In many cases, this central programme approach has promoted an evolution of technologies resulting in optimisation of simple remote tooling. In other instances, collaboration with suppliers is producing state-of-the-art technologies to tackle more unique challenges including, most recently, early demonstration of laser cutting for reactor vessel dismantling. This paper will outline the programme management approach and the scalable optioneering process Magnox employs to select the engineered controls and technologies used in decommissioning and waste management activities. It will provide an overview of the broad range of remote tooling and robotics deployed across the 12 Magnox sites by exploring some specific examples. Lastly, the paper will summarise key lessons learned in the now extensive history of selecting and using remote technologies to decommission Magnox nuclear facilities. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2021

Privacy-Preserving Knowledge Transfer with Bootstrap Aggregation of Teacher Ensembles

There is a need to transfer knowledge among institutions and organizations to save effort in annotation and labeling or in enhancing task performance. However, knowledge transfer is difficult because of restrictions that are in place to ensure data security and privacy. Institutions are not allowed to exchange data or perform any activity that may expose personal information. With the leverage of a differential privacy algorithm in a high-performance computing environment, we propose a new training protocol, Bootstrap Aggregation of Teacher Ensembles (BATE), which is applicable to various types of machine learning models. The BATE algorithm is based on and provides enhancements to the PATE algorithm, maintaining competitive task performance scores on complex datasets with underrepresented class labels.We conducted a proof-of-the-concept study of the information extraction from cancer pathology report data from four cancer registries and performed comparisons between four scenarios: no collaboration, no privacy-preserving collaboration, the PATE algorithm, and the proposed BATE algorithm. The results showed that the BATE algorithm maintained competitive macro-averaged F1 scores, demonstrating that the suggested algorithm is an effective yet privacy-preserving method for machine learning and deep learning solutions.

Yoon, Hong-Jun↗
DOE OSTI2022

Advances in Machine and Deep Learning for Modeling and Real-time Detection of Multi-Messenger Sources

We live in momentous times. The science community is empowered with an arsenal of cosmic messengers to study the universe in unprecedented detail. Gravitational waves, electromagnetic waves, neutrinos, and cosmic rays cover a wide range of wavelengths and timescales. Combining and processing these datasets that vary in volume, speed, and dimensionality requires new modes of instrument coordination, funding, and international collaboration with a specialized human and technological infrastructure. In tandem with the advent of large-scale scientific facilities, the last decade has experienced an unprecedented transformation in computing and signal-processing algorithms. The combination of graphics processing units, deep learning, and the availability of open source, high-quality datasets has powered the rise of artificial intelligence. This digital revolution now powers a multibillion dollar industry, with far-reaching implications in technology and society. In this chapter, we describe pioneering efforts to adapt artificial intelligence algorithms to address computational grand challenges in multi-messenger astrophysics. We review the rapid evolution of these disruptive algorithms, from the first class of algorithms introduced in early 2017 to the sophisticated algorithms that now incorporate domain expertise in their architectural design and optimization schemes. We discuss the importance of scientific visualization and extreme-scale computing in reducing time-to-insight and obtaining new knowledge from the interplay between models and data.

Artificial Intelligence↗
DOE OSTI2025

Understanding EV Charging Pain Points Through Deep Learning Analysis

Current and potential electric vehicle (EV) owners express concerns about the charging infrastructure, mentioning non-functional chargers, prolonged charging times, inconvenient charger locations, long wait times, and high costs as major barriers. Addressing these issues often requires analyzing actual vehicle charging data, which is typically proprietary and inconsistent due to diverse standards and protocols. To understand and improve the EV charging experience, customer reviews are typically used to identify common customer pain points (CPPs). However, there is not a comprehensive method to map customer reviews to a standardized set of CPPs. In collaboration with the National Charging Experience (ChargeX) Consortium, this study bridges these gaps by proposing a Systematic Categorization and Analysis of Large-scale EV-charging Reviews (SCALER) framework. SCALER is an integrated, deep learning framework that segments, actively labels, analyzes, and classifies EV charging customer reviews into six CPP categories. To test its effectiveness, we used SCALER to analyze over 72,000 reviews from customers charging various EV models on different networks across the United States. SCALER achieves a classification accuracy of 92.5%, with an F1 score exceeding 85.7%. By demonstrating real-world applications of SCALER, we enhance the industry’s ability to understand and address CPPs to improve the EV charging experience.

29 - ENERGY PLANNING, POLICY AND ECONOMY↗
DOE OSTI2022

NNMC–DP Continuing Education Summary [Slides]

DP technicians voiced a desire for more development and learning opportunities. DP management and technician partners identified the opportunity to expand the knowledge of the technicians through employee development to strengthen the collaboration.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2023

Use of Stable Mercury Isotopes to Assess Mercury and Methylmercury Transformation and Transport across Critical Interfaces from the Molecular to the Watershed Scale (Final Report)

This project titled “Use of Stable Mercury Isotopes to Assess Mercury and Methylmercury Transformation and Transport across Critical Interfaces from the Molecular to the Watershed Scale” represents a collaborative effort between the University of Michigan (Jason Demers, PI) and Oak Ridge National Laboratory (Scott Brooks, co-I). Much has been learned about mercury (Hg) cycling in stream ecosystems, and East Fork Poplar Creek (EFPC) in particular, through decades of previous research. Nevertheless, some of the most fundamental questions regarding the sources of bioavailable Hg and its transformation to toxic methylmercury (MeHg) have remained unanswered. These fundamental questions include: (1) what are the sources and biogeochemical processes that lead to the input of dissolved Hg to stream water across critical subsurface interfaces within stream ecosystems, and EFPC in particular? and (2) what are the sources and biogeochemical processes that control the production and fate of bioaccumulative MeHg within stream ecosystems, and in EFPC in particular? To address these fundamental questions, our project aimed to couple laboratory experiments and field observations, both utilizing natural abundance Hg stable isotope techniques, to identify the processes responsible for generating mobile, bioavailable dissolved Hg from recalcitrant legacy sources within critical subsurface zones (e.g., streambed hyporheic zone, riparian floodplain subsurface). We used the isotopic signature of this bioavailable dissolved Hg to track its mobilization across these critical interfaces in order to link diffuse subsurface sources of dissolved Hg with increases in surface water dissolved Hg flux measured at the watershed scale. Additionally, our research aimed to determine the isotopic composition of MeHg within these same critical subsurface zones. We directly assessed the isotopic composition of MeHg within biota in order to gain insight into which subsurface sources of inorganic Hg and toxic MeHg are available for bioaccumulation within the EFPC ecosystem. Net fluxes of dissolved Hg along the flow path of EFPC were shown to vary spatially and temporally. In the Upper EFPC, within the Y12 boundary, stream water flux of dissolved Hg consistently decreased between the outfall and the downstream boundary of Y12 (57% ± 29%, 1SD). Within the Upper EFPC, an assessment of Hg isotopic composition suggested that losses were strongly reaction-driven, although isotopic diagnostics did not conform to any known processes. Downstream of Y12, in the upper reach of the Lower EFPC, dissolved Hg fluxes tended to increase during the dormant season (net gain of 11-120%), and decrease during the growing season (net loss of 23% +/- 18%, 1SD). In the downstream-most reach of Lower EFPC, dissolved Hg fluxes increased by 12-108% in 9 out of 10 monthly assessments. Overall, diffuse fluxes from the non-Y12 watershed accounted for 34% (+/- 17%, 1SD) of all dissolved Hg exported during base flow. Within Lower EFPC, an assessment of Hg isotopic composition was consistent with the contribution of diffuse Hg inputs from high-concentration hotspots within riparian floodplains and streambed hyporheic pore water. To investigate remobilization of recalcitrant Hg from legacy sediment sources, we developed procedures that coupled isotopic analysis with sequential extractions of streambed sediment. We found that the proportion of weakly-bound Hg within EFPC streambed sediment was relatively small, but could still account for a large proportion of the annual flux of dissolved Hg from EFPC. These sequential extractions also showed that this weakly-bound Hg fraction could be replenished from the much larger fraction of recalcitrant Hg in sediment. The isotopic composition of these weakly-bound and remobilized recalcitrant Hg fractions within the sediment was consistent with high-concentration dissolved Hg hotspots within hyporheic pore water. Thus, this research provided novel evidence that legacy mercury sources within streambed sediment could provide an ongoing contribution of dissolved Hg to surface waters. Finally, we developed new methods for the direct determination of the MeHg isotopic composition of organisms, which allowed a more direct evaluation of inorganic Hg and MeHg sources accumulating in the food web. We found that fish and aquatic invertebrates in both EFPC and a regional background site obtained inorganic Hg and MeHg from multiple isotopically distinct sources, including sediment, suspended particulates, and periphyton. Photodemethylation was found to be an important reaction influencing MeHg dynamics at both sites. However, the balance of microbial methylation and demethylation processes differed between the two streams, with fractionation resulting from methylation and demethylation processes being relatively in balance within the regional background site, whereas microbial methylation appeared to be dominant over microbial demethylation within the EFPC ecosystem. Broadly, the application of Hg isotopic analysis in this study led to numerous novel insights regarding the biogeochemical cycling of Hg in stream ecosystems. This research project promoted the development of two new approaches, including the coupling of sequential extractions with Hg isotopic analysis to assess remobilization of recalcitrant Hg within sediments, and a new method for the isotopic analysis of MeHg isolated from environmental samples. Both of these efforts represent advances in capacity for the field of mercury isotopic analysis and environmental assessment. Overall, this study demonstrates that the application of Hg stable isotope techniques continues to provide new insights into the biogeochemical cycling of Hg in complex aquatic environments, both within the EFPC and beyond.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Strategies and Good Practices to Support Robust Stakeholder Engagement in Multi-Sector Energy Transition Planning

This document discusses strategies and good practices to facilitate diverse stakeholder engagement and inclusive decision-making within the context of complex energy system modernization studies and multi-sector energy transition planning, drawing upon the knowledge and expertise of numerous USAID, NREL, and other DOE laboratory practitioners with international and domestic energy planning experience. Examples of multi-sector energy transition analyses from the United States are included in this document to draw out relevant lessons learned and experience for the USAID-NREL Partnership - strategies and good practices will be updated over time as new insights emerge through collaboration with international partners and stakeholders. Considerations included within this document can also be used to support stakeholder engagement at the sector level (e.g., for power, transport, buildings), in addition to integrated approaches.

ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2021

Using machine learning to improve efficiency and accuracy of burnup measurements at PBR reactors [Slides]

The outline of the slides include: Motivations of the work; Modeling and simulation; Machine learning model; Results and comparison study with linear regression; and Conclusions. This work was done to help PBR designers and operators understand the burnup measurement better. We look forward to discussing the results in detail with industrial collaborators.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2025

CAFE AU LAIT: Compute-Aware Federated Augmented Low-Rank AI Training

Federated finetuning is crucial for unlocking the knowledge embedded in pretrained Large Language Models (LLMs) when data are geographically distributed across clients. Unlike finetuning with data from a single institution, federated finetuning allows collaboration across multiple institutions, enabling the utilization of diverse and decentralized datasets while preserving data privacy. Given the high computing costs of LLM training and the emphasis on energy efficiency in Federated Learning (FL), Low-Rank Adaptation (LoRA) has emerged as a widely adopted algorithm due to its significantly reduced number of trainable parameters. However, this assumes that all data silos have the necessary computing resources to compute local updates of LLMs. Nevertheless, in practice, the computing resources across clients are highly heterogeneous: while some may have access to hundreds of GPUs, others might have limited or no GPU access. Recently, federated finetuning using synthetic data has been proposed, allowing clients to participate in a collaborative training run without training LLMs locally. However, our experimental results reveal a performance gap between models trained using synthetic data and those trained using local updates. Motivated by the observed heterogeneity in computing resources and the performance gap, we propose a novel two-stage algorithm that leverages the storage and computing capabilities of a strong server. In the first stage, under the coordination of the strong server, clients with limited computing resources collaborate to generate synthetic data, which is transferred to and stored on the strong server. In the second stage, the strong server uses this synthetic data on behalf of the resource-constrained clients to perform federated LoRA finetuning alongside clients with sufficient computing resources. This approach ensures that all clients can participate in the finetuning process. Experimental results demonstrate that incorporating local updates from even a small fraction of clients improves performance compared to using synthetic data for all clients. Furthermore, we incorporate the Gaussian mechanism in both stages to guarantee client-level differential privacy.

Wang, Jiayi [ORNL]↗
DOE OSTI2021

Crowdsourcing Global Perspectives in Ecology Using Social Media

Transparent, open, and reproducible research is still far from routine, and the full potential of open science has not yet been realized. Crowdsourcing–defined as the usage of a flexible open call to a heterogeneous group of individuals to recruit volunteers for a task –is an emerging scientific model that encourages larger and more outwardly transparent collaborations. While crowdsourcing, particularly through citizen- or community-based science, has been increasing over the last decade in ecological research, it remains infrequently used as a means of generating scientific knowledge in comparison to more traditional approaches. We explored a new implementation of crowdsourcing by using an open call on social media to assess its utility to address fundamental ecological questions. We specifically focused on pervasive challenges in predicting, mitigating, and understanding the consequences of disturbances. In this paper, we briefly review open science concepts and their benefits, and then focus on the new methods we used to generate a scientific publication. We share our approach, lessons learned, and potential pathways forward for expanding open science. Our model is based on the beliefs that social media can be a powerful tool for idea generation and that open collaborative writing processes can enhance scientific outcomes. We structured the project in five phases: (1) draft idea generation, (2) leadership team recruitment and project development, (3) open collaborator recruitment via social media, (4) iterative paper development, and (5) final editing, authorship assignment, and submission by the leadership team. We observed benefits including: facilitating connections between unusual networks of scientists, providing opportunities for early career and underrepresented groups of scientists, and rapid knowledge exchange that generated multidisciplinary ideas. We also identified areas for improvement, highlighting biases in the individuals that self-selected participation and acknowledging remaining barriers to contributing new or incompletely formed ideas into a public document. While shifting scientific paradigms to completely open science is a long-term process, our hope in publishing this work is to encourage others to build upon and improve our efforts in new and creative ways.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Business Models for Scaling Demand Flexibility Volume III – Stakeholder ecosystem management challenges, and lessons learned from U.S. programs

Load growth at the grid edge is driving increased attention to the distribution system and its ability to enable customer technology adoption in an affordable and timely manner. Key industry stakeholders, including electric utilities and regulators, can benefit from strategies to manage and balance customer needs with infrastructure investments, such as demand flexibility. This report focuses on demand flexibility—the ability to reduce, shift, shed, generate, or modulate loads in response to building and grid needs—to reduce the need for costly grid upgrades by deferring investment needs and increase system reliability by shifting electricity usage during periods of high risk. Specifically, we focus on the emerging characteristics of business models for demand flexibility as a framework to understand how demand flexibility programs generate value. In this report, we focus on demand flexibility program stakeholder ecosystem management strategies, which provide information on value delivery and describe how program implementers leverage partner capabilities and collaborate to deliver customer and grid benefits. This report discusses the role of stakeholder ecosystem management in demand flexibility programs, identifies existing challenges to effective stakeholder management, and describes lessons learned. This report is part of a series that includes reports on value propositions, customer relationship management strategies, and program life cycle.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2022

Combining chemistry and protein engineering for new-to-nature biocatalysis

Biocatalysis, the application of enzymes to solve synthetic problems of human import, has blossomed into a powerful technology for chemical innovation. In the past decade, a threefold partnership, where nature provides blueprints for enzymatic catalysis, chemists introduce innovative activity modes with abiological substrates, and protein engineers develop new tools and algorithms to tune and improve enzymatic function, has unveiled the frontier of new-to-nature enzyme catalysis. In this Perspective, we highlight examples of interdisciplinary studies, which have helped to expand the scope of biocatalysis, including concepts of enzymatic versatility explored through the lens of biomimicry, to achieve activities and selectivities beyond those currently possible with chemocatalysis. We indicate how modern tools, such as directed evolution, computational protein design and machine learning-based protein engineering methods, have already impacted and will continue to influence enzyme engineering for new abiological transformations. As a result, a sustained collaborative effort across disciplines is anticipated to spur further advances in biocatalysis in the coming years.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2018

Final Report for CRADA SC16/01880.00

In this project, Sandia National Laboratories (SNL) and Rigetti Quantum Computing (RQC) collaborated to test and evaluate (T&E) an 8-qubit quantum information processor developed by RQC. SNL provided quantum circuits and data analysis protocols to benchmark the performance of the device. These protocols included gate-set tomography (GST) on 1- and 2-qubit subsystems, as well as more advanced protocols for characterizing crosstalk and drift. RQC provided high- and low-level access to the 8-qubit device, via direct collaboration: SNL was able to specify T&E experiments as lists of quantum circuits, delivered to RQC via email, and RQC then ran these experiments and delivered the resulting data back to SNL for analysis. The collaboration characterized the performance of all 1-qubit gates on all 8 qubits, and a sample 2-qubit gate on one pair of qubits. Extensive investigation of crosstalk produced nondefinitive results, and revealed significant drift that interfered with crosstalk detection. Drift was analyzed. As a result of this collaboration, SNL was able to test protocols for crosstalk and drift, and improve them in light of experimental data, while RQC was able to learn about the performance of their device and improve it by debugging.

Blume-Kohout, Robin↗
DOE OSTI2023

The LSST AGN Data Challenge: Selection Methods

Abstract Development of the Rubin Observatory Legacy Survey of Space and Time (LSST) includes a series of Data Challenges (DCs) arranged by various LSST Scientific Collaborations that are taking place during the project's preoperational phase. The AGN Science Collaboration Data Challenge (AGNSC-DC) is a partial prototype of the expected LSST data on active galactic nuclei (AGNs), aimed at validating machine learning approaches for AGN selection and characterization in large surveys like LSST. The AGNSC-DC took place in 2021, focusing on accuracy, robustness, and scalability. The training and the blinded data sets were constructed to mimic the future LSST release catalogs using the data from the Sloan Digital Sky Survey Stripe 82 region and the XMM-Newton Large Scale Structure Survey region. Data features were divided into astrometry, photometry, color, morphology, redshift, and class label with the addition of variability features and images. We present the results of four submitted solutions to DCs using both classical and machine learning methods. We systematically test the performance of supervised models (support vector machine, random forest, extreme gradient boosting, artificial neural network, convolutional neural network) and unsupervised ones (deep embedding clustering) when applied to the problem of classifying/clustering sources as stars, galaxies, or AGNs. We obtained classification accuracy of 97.5% for supervised models and clustering accuracy of 96.0% for unsupervised ones and 95.0% with a classic approach for a blinded data set. We find that variability features significantly improve the accuracy of the trained models, and correlation analysis among different bands enables a fast and inexpensive first-order selection of quasar candidates.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2025

A New Framework for Interstellar Medium Emission Line Models: Connecting Multiscale Simulations across Cosmological Volumes

The James Webb Space Telescope (JWST) and Atacama Large Millimeter/submillimeter Array have detected emission lines from the ionized interstellar medium (ISM) in some of the first galaxies at z ≳ 6. These measurements present an opportunity to better understand galaxy assembly histories and may allow important tests of state-of-the-art galaxy formation simulations. It is challenging, however, to model these lines in their proper cosmological context. In order to meet this challenge, we introduce a novel subgrid line emission modeling framework. The framework uses the high-z zoom-in simulation suite from the Feedback in Realistic Environments (FIRE) collaboration. The line emission signals from H II regions within each simulated FIRE galaxy are modeled using the semianalytic HIIL INES code. A machine learning approach is then used to determine the conditional probability distribution for the line luminosity to stellar-mass ratio from the H II regions around each simulated stellar particle. This conditional probability distribution can then be applied to predict the line luminosities around stellar particles in lower-resolution, yet larger volume cosmological simulations. As an example, we apply this approach to the IllustrisTNG simulations at z = 6. The resulting predictions for the [O II ], [O III ], and Balmer line luminosities as a function of star formation rate agree well with current observations. Our predictions differ, however, from related works in the literature, which lack detailed subgrid ISM models. This highlights the importance of our multiscale simulation modeling framework. Finally, we provide forecasts for future line luminosity function measurements from the JWST and quantify the cosmic variance in such surveys.

(ISM:) H II regions↗