Search NASA⌕ Search

SEARCH · Search NASA

Results for “scientific training”

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 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

DOE BSSD Performance Management Metrics Report Q2

The vision of the National Microbiome Data Collaborative (NMDC) centers on the concept of connecting data, people, and ideas to advance microbiome innovation and discovery. Building data infrastructure, while key to NMDC’s ability to execute on our vision, can only go so far in creating scientific impact. By fostering strong community partnerships and developing a set of robust community outreach and training programs, we are able to turn our products – the Submission Portal, NMDC EDGE, and the Data Portal – into tools that empower the scientific community. Our multi-pronged community building approach spans individual researchers, research teams, consortia and scientific societies, and institutions and federal agencies. To foster a collaborative and inclusive community-centered environment, we have identified three strategic objectives to promote an inclusive and connected community: (1) recognize and support the diverse research needs and perspectives of the microbiome research community; (2) promote best practices across the microbiome community, from researchers to funders, through community-driven practices (FAIR, CARE, and TRUST); and (3) build a microbiome ecosystem that enables scientific discovery and innovation across stakeholders. These strategic objectives allow our team to focus on impact across a diverse range of activities, from launching the American Society for Microbiology (ASM) Microbiome Data Prize to supporting the Ambassador and Champions programs fostering learning and building a collaborative network. We broadly communicate our work through social media (X/Twitter, LinkedIn, and Instagram), The Microbiome Standard (our quarterly newsletter), and Annual Reports. All our work is underpinned by a strong commitment to diversity, equity, and inclusion as articulated in our Action Plan that tracks progress towards key metrics. A core component of our engagement strategy is user research. User research ensures the Submission Portal, NMDC EDGE, Data Portal, and the new Field Notes mobile app are designed with and for the scientific community. Our user research efforts consist of asking researchers exploratory questions to collect information on researcher priorities, methodologies, and perceptions to ensure that we are aware of the current state of microbiome research. Our usability testing provides researchers with prototypes or test environments of the NMDC products, and we capture valuable information on how users interact with the products to make improvements. Given the diverse nature of microbiome work, we acknowledge that we are not aware of all pressing data challenges and thus rely on the research community to help us identify the most important issues to prioritize. To date, we have conducted 24 interviews and one beta-testing call with 10 participants across all NMDC products, which have generated 321 insights and 120 action items. Herein, we describe the ways we engage with the microbiome research community to advance the NMDC mission.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Ecosystems for Scientific Computing in the Age of AI

Scientific computing is at an inflection point. Artificial intelligence (AI) is reshaping how scientific software is developed, how teams collaborate, how projects are governed, and how the next generation is trained. Drawing on insights from a 2025 workshop report, this article argues that the future of discovery will depend on agile, robust ecosystems built through socio-technical co-design—the intentional integration of technical and human systems. This perspective is essential for ensuring that future scientific computing remains trustworthy, sustainable, and scalable. It combines advances in AI, high-performance computing, and software with new models for cross-disciplinary collaboration, education, and workforce development. Key recommendations include building modular, trustworthy AI-enabled software ecosystems; enabling teams to integrate AI into scientific workflows while preserving human creativity, integrity, and rigor; and developing adaptive training pathways that keep pace with rapid technological change. By sharing these perspectives, we hope to stimulate broader community dialogue and encourage coordinated action.

AI↗
DOE OSTI2026

Final Technical Report for U.S.-Japan Hadronic Physics Exchange Program for Studies of Hadron Structure and QCD

Nuclear physics explores the fundamental properties of matter -- how protons and neutrons emerge as quantum systems of elementary particles, how they form the atomic nuclei, and how they give rise to the wide variety of phenomena and applications at biological, technical, and astronomical scales. It is a global scientific effort centered around large-scale experimental user facilities (particle accelerators and detectors), advanced theoretical methods and concepts, and computational techniques and resources. Exchange of knowledge and ideas, scientific collaboration, and workforce development on a global scale are essential for the future of the field. The nuclear physics program envisaged in the 2023 DOE/NSF NSAC Long-Range Plan and pursued at the U.S. National Labs has strong synergies with programs at other facilities worldwide and will realize significant benefits from international collaboration. Nuclear physics is also recognized for promoting international cooperation in the broadest sense through joint construction and operation of experimental equipment, personal contacts between scientists, and education and training. The U.S.-Japan Hadronic Physics Exchange Program (USJPHE) supported collaborative scientific research in hadronic physics and quantum chromodynamics. USJHPE focused on subject areas related to the programs at current and future experimental facilities in the U.S.\ and Japan and supported both experimental and theoretical studies. USJHPE particularly aimed to realize synergies between the hadronic physics programs at Jefferson Lab 12 GeV and J-PARC resulting from the complementarity of electromagnetic and hadronic probes in the multi-GeV energy range. Subject areas of common interest included the quark-gluon structure of hadrons and nuclei, meson and baryon spectroscopy, strangeness and hypernuclear physics, and other related topics. USJHPE also supported research in hadronic physics and nuclear-physics-enabled tests of fundamental symmetries related to the programs at Brookhaven National Lab, Fermilab, KEK, Spring-8, and university-based facilities in the U.S. and Japan. USJHPE especially promoted collaboration between the U.S. and Japanese nuclear physics communities in developing the physics program and instrumentation for the future Electron-Ion Collider. USJHPE was intended to provide travel grants to U.S.-based scientists (primary institutional affiliation with a U.S.\ university, national laboratory, or other research center) to visit Japanese institutions and conduct collaborative research there. The program supported senior researchers, postdoctoral fellows, and students. Continuing the setup of the preceding grant period, J-PARC served as the Japanese “hub” for U.S. physicists for short- and long-term visits, and JLab served as the corresponding U.S. “hub”. The program was officially managed through the U. of Connecticut in Storrs, CT. Support for Japanese physicists visiting the U.S. was provided through funds from Japanese funding agencies. The USJHPE program promoted the scientific exchange and the collaborative spirit in hadronic physics between the two countries.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2022

Unsupervised quantum circuit learning in high energy physics

Unsupervised training of generative models is a machine learning task that has many applications in scientific computing. Here, in this work, we evaluate the efficacy of using quantum circuit-based generative models to generate synthetic data of high energy physics processes. We use nonadversarial, gradient-based training of quantum circuit Born machines to generate joint distributions over two and three variables.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

2020 Budget Request for the DOE Computational Science Graduate Fellowship (CSGF) Grant

The Department of Energy Computational Science Graduate Fellowship (DOE CSGF) is essential for addressing the increasingly complex national workforce demands stemming from the growth of computational science and engineering challenges. Computational science and engineering (CSE) takes a multidisciplinary approach that utilizes scientific computing to tackle practical problems and provide technical tools across the spectrum of scientific discovery. The DOE CSGF specifically highlights high-performance computing (HPC) as a critical enabling technology in CSE, driving advancements in science and engineering that are vital to both the DOE and the broader economy. Over the past half-century, HPC has been an essential tool for DOE’s success. During this period, important missions, such as nuclear stockpile stewardship, have turned to HPC as an essential technology. Entire science disciplines have been transformed through the augmentation of scientific observation via HPC. At government laboratories, academic institutions, and in industry, DOE CSGF alumni are helping push traditional HPC boundaries while contributing to discoveries in high-energy physics, quantum information systems, fusion-reactor design, machine learning, additive manufacturing, nano materials for next-generation batteries and transistors, and advanced nuclear reactor modeling. In addition, HPC is used to address national health needs that will eventually point to cures both by helping cancer researchers manage and analyze huge troves of data, by simulating biological mechanisms, and by accelerating drug development. A 2023 report from the ASCAC Subcommittee on American Competitiveness and Innovation to the ASCR office, “Can the United States Maintain Its Leadership in High-Performance Computing?” says of the Program, “The CSGF program provides a barometer for disciplines that will be of interest to future DOE computing. Computational biology, machine learning, and quantum computing are among the subjects that began to swell in the ranks of CSGF applicants before the labs were hiring as high a percentage of employees in these categories.” The explosion of scientific and technological data has heightened the demand for advanced high-performance computing (HPC) to transform these data into meaningful scientific insights. As access to vast amounts of data increases, the fields of Machine Learning and Artificial Intelligence are experiencing a resurgence, enhancing the established practices of computational modeling and simulation. In its September 2020 subcommittee report on "AI/ML, Data Intensive Science, and High-Performance Computing," the DOE Advanced Scientific Computing Advisory Committee (ASCAC) specifically called for a fellowship program to train computational and data scientists to address exascale and data-intensive computing challenges. This integration of empirical and theoretical modeling will increasingly guide federal policymakers in making decisions that impact American society and future generations. It demands a workforce of highly skilled and intellectually agile computational scientists capable of navigating the rapid advancements in scientific computing within the DOE National Laboratory research environment. The DOE CSGF program has consistently addressed this critical need.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

DNTTD (Distributed Non-Negative Tensor Train Decomposition)

The era of exascale computing opens new venues for innovations and discoveries in many scientific, engineering, and commercial fields. However, with the exa flops also come the extra-large high-dimensional data generated by high performance computing. High-dimensional data is presented as multidimensional arrays, aka tensors. The presence of latent (not directly observable) structures in the tensor allows a unique representation and compression of the data by classical tensor factorization techniques. However, the classical tensor methods are not always stable or they can be exponential in their memory requirements, which makes them not suitable for high-dimensional tensors. Tensor train (TT) is a state-of-the-art tensor network introduced for factorization of high-dimensional tensors. TT transforms the initial high-dimensional tensor in a network of three-dimensional tensors that requires only a linear storage. Many real-world data, such as, density, temperature, population, probability, etc., are non-negative and for an easy interpretation, the algorithms preserving non-negativity are preferred. Here, we introduce a distributed non-negative tensor-train and demonstrate its scalability and the compression on synthetic and real world big datasets.

Bhattarai, Manish↗
DOE OSTI2020

Usable Data Abstractions for Next-Generation Scientific Workflows

Data- and computationally-intensive scientific research, such as numerical simulations and inversions or the training of large neural networks in machine learning applications, that are well suited for HPC environments also often require expert insight and evaluation throughout the computation which can be greatly facilitated with the use of interactive computing tools, such as those in the Jupyter ecosystem. HPC workflows and interactive workflows are typically treated as orthogonal, however, the next generation of research will require both. The first challenge we face in this project is thus designing the right level of abstractions to allow interactive capabilities in the JupyterLab environment to allow the working scientist to flexibly explore and query their data at multiple levels, with a minimal amount of customization required of the underlying optimized codes. In addition to these questions regarding the high-level representation of data for interactive use in HPC, we tackled two additional issues that are part of the entire lifecycle of research and that become particularly acute in HPC contexts: how to improve the experience of interfacing with the HPC system's scheduling environment for a scientist focused on exploratory questions, and how can that scientist then best share the results of their work with others in a self-contained, reproducible manner.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Measuring Vibronic Coupling and Ultrafast Charge Delocalization on Nanocrystal Surfaces Using Ligand-Specific Vibrational Probes

The primary goal of this proposal was to measure and quantify how vibronic ligand-exciton interactions at the inorganic-organic interface of semiconductor nanocrystals (NCs) govern femtosecond non-radiative relaxation and charge delocalization mechanisms as a function of nanocrystal and ligand structures – including NC size, shape, and ligand identity. Molecular-scale chemical and physical properties and interactions of surface ligands dictate nanocrystal exciton dynamics on the femto-to-microsecond timescales. Improvements in synthesis, experimental, and theoretical characterizations have enhanced our understanding of NC photophysics. However, quantitative descriptions of complex ligand-exciton interactions at the NC surface have remained elusive, which limits elucidation of relaxation mechanisms that determine overall carrier transfer dynamics at the exciton-ligand interface. Thus, the central motivation for this proposal was to fill a significant knowledge gap for judicious selection of ligand-functionalized nanocrystals for targeted isolation and capture of charge carriers to drive selective organic photocatalytic transformations in NC photocatalysts and NC-assembled photonic materials. Our unique two-pronged experimental approach combining nanocrystal synthesis and ligand-specific nonlinear spectroscopy has resulted in (i) four peer-reviewed publications; (ii) 12+ invited and contributed posters and talks at national and international scientific meetings and seminars, (iii) two completed PhD theses, (iv) training of two postdoctoral researchers, three PhD students, and two undergraduate students, and (v) two manuscripts under preparation.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Focused Ion Beam Tomography of Alloy 617 Corroded in Molten Chloride Salt

Materials qualification of reactor structural materials is a critical step in rapid implementation of advanced nuclear reactor technologies, particularly to assess the corrosion performance in these designs. Accelerated qualification of reactor structural materials requires incorporating powerful computational toolsets, such as phase field modelling in the Multiphysics Object-Oriented Simulation Environment (MOOSE) framework, to predict the evolution of structural materials due to corrosion. Accordingly, computational toolsets will require experimental data generated at appropriate length scales to validate accuracy. Focused ion beam (FIB) provides a high degree of control over manipulation of materials for analytical purposes, including capturing data on the evolution in the microstructure and elemental composition of materials at the mesoscale, an appropriate length scale for phase field modelling of intergranular diffusion phenomena using the MOOSE framework. For instance, the FEI Helios G4 UX dual beam plasma FIB microscope at the Irradiated Materials Characterization Laboratory (IMCL) is capable of backscatter diffraction (EBSD) and energy-dispersive x-ray spectroscopy (EDS) documenting the evolution in the microstructure and elemental composition, respectively. The Helios can perform EDS and EBSD three-dimensionally (3D) using tomography, which is then combined using different software packages to visualize 3D volumes correlating elemental composition to microstructural data. The purpose of this investigation was to develop a streamlined characterization and data processing workflow for 3D tomography studies on the FEI Helios G4 plasma FIB. The investigation is segmented into three parts: 1) Optimizing the data collection workflow, 2) identifying appropriate data processing and visualization software (i.e. DREAM.3D, MIPAR, and VGStudioMax), and 3) establishing an infrastructure for public release. The optimization of the data collection workflow is in collaboration with members of the U220 department to setup formal training on the tomography operation of the G4, through ThermoFisher Scientific, and exploring DREAM.3D, MIPAR, and VGStudioMax data processing/visualization software packages. VGStudioMax currently demonstrates the most promise for future use. Optimization of the data collection and processing workflow is still ongoing. A collaboration with INL High Performance Computing (HPC) established an open-source license for expediting the public release of FIB tomography datasets through HPC. FIB tomography data generated by the G4 will provide comprehensive data for validating 3D phase field mesoscale modelling tools within the MOOSE framework for accelerated qualification of reactor structural materials.

Copeland-Johnson, Trishelle↗
DOE OSTI2025

Optimizing the optimizer for physics-informed neural networks and Kolmogorov-Arnold networks

Physics-Informed Neural Networks (PINNs) have revolutionized the computation of PDE solutions by integrating partial differential equations (PDEs) into the neural network’s training process as soft constraints, becoming an important component of the scientific machine learning (SciML) ecosystem. More recently, physics-informed Kolmogorv-Arnold networks (PIKANs) have also shown to be effective and comparable in accuracy with PINNs. In their current implementation, both PINNs and PIKANs are mainly optimized using first-order methods like Adam, as well as quasi-Newton methods such as BFGS and its low-memory variant, L-BFGS. However, these optimizers often struggle with highly nonlinear and non-convex loss landscapes, leading to challenges such as slow convergence, local minima entrapment, and (non)degenerate saddle points. In this study, we investigate the performance of Self- Scaled BFGS (SSBFGS), Self-Scaled Broyden (SSBroyden) methods and other advanced quasi-Newton schemes, including BFGS and L-BFGS with different line search strategies. These methods dynamically rescale updates based on historical gradient information, thus enhancing training efficiency and accuracy. We systematically compare these optimizers – using both PINNs and PIKANs – on key challenging PDEs, including the Burgers, Allen-Cahn, Kuramoto-Sivashinsky, Ginzburg-Landau, and Stokes equations. Additionally, we evaluate the performance of SSBFGS and SSBroyden for Deep Operator Network (DeepONet) architectures, demonstrating their effectiveness for data-driven operator learning. Our findings provide state-of-the-art results with orders-of-magnitude accuracy improvements without the use of adaptive weights or any other enhancements typically employed in PINNs. More broadly, our work reveal insights into the effectiveness of quasi-Newton optimization strategies in significantly improving the convergence and accurate generalization of PINNs and PIKANs.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Making Invisible Visible: Data-Driven Seismic Inversion With Spatio-Temporally Constrained Data Augmentation

Deep learning and data-driven approaches have shown great potential in scientific domains. The promise of data-driven techniques relies on the availability of a large volume of high-quality training datasets. Due to the high cost of obtaining data through expensive physical experiments, instruments, and simulations, data augmentation techniques for scientific applications have emerged as a new direction for obtaining scientific data recently. However, existing data augmentation techniques originating from computer vision yield physically unacceptable data samples that are not helpful for the domain problems that we are interested in. In this article, we develop new data augmentation techniques based on convolutional neural networks. Specifically, our generative models leverage different physics knowledge (such as governing equations, observable perception, and physics phenomena) to improve the quality of the synthetic data. To validate the effectiveness of our data augmentation techniques, we apply them to solve a subsurface seismic full-waveform inversion using simulated CO 2 leakage data. Our interest is to invert for subsurface velocity models associated with very small CO 2 leakage. We validate the performance of our methods using comprehensive numerical tests. Here via comparison and analysis, we show that data-driven seismic imaging can be significantly enhanced by using our data augmentation techniques. Particularly, the imaging quality has been improved by 15% in test scenarios of general-sized leakage and 17% in small-sized leakage when using an augmented training set obtained with our techniques.

58 GEOSCIENCES↗
DOE OSTI2025

FM4NPP: A Scaling Foundation Model for Nuclear and Particle Physics

Large language models have revolutionized artificial intelligence by enabling large, generalizable models trained through self-supervision. This paradigm has inspired the development of scientific foundation models (FMs). However, applying this capability to experimental particle physics is challenging due to the sparse, spatially distributed nature of detector data, which differs dramatically from natural language. This work addresses if an FM for particle physics can scale and generalize across diverse tasks. We introduce a new dataset with more than 11 million particle collision events and a suite of downstream tasks and labeled data for evaluation. We propose a novel self-supervised training method for detector data and demonstrate its neural scalability with models that feature up to 188 million parameters. With frozen weights and task-specific adapters, this FM consistently outperforms baseline models across all downstream tasks. The performance also exhibits robust data-efficient adaptation. Further analysis reveals that the representations extracted by the FM are task-agnostic but can be specialized via a single linear mapping for different downstream tasks.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2026

SetGo: Metadata Readiness for Scientific AI Datasets

Scientific datasets intended for AI use require both computational readiness for model training and metadata readiness for discovery, sharing, and reuse. The Readiness Engine for Data Integration (REDI) addresses computational readiness, but no corresponding tool evaluates whether a dataset’s metadata are sufficiently complete, governed, and standards-compliant for publication and agent-based consumption. Existing FAIR assessors operate only on published repository records, and no single system covers FAIR compliance, licensing, provenance, governance, reproducibility, and catalog readiness together. We present SetGo, an open-source Python toolkit that assesses and repairs metadata readiness across these six dimensions before a dataset is published or archived. Applied to four scientific corpora, SetGo surfaces deficiencies that general-purpose tools do not detect: ERA5 climate metadata scores 4% on ACDD 1.3 compliance; materials datasets fail OPTIMADE species-definition requirements; and PDB-derived proteomics data carries licensing terms incompatible with standard SPDX identifiers. Guided enrichment raises overall FAIR scores from 52–57% to 81–91%, and a single setgo publish command pushes to Hugging Face Hub, CKAN, or OpenMetadata with ML Commons Croissant 1.0 metadata sidecars. To support interactive and automated workflows, SetGo integrates with coding agents powered by large language models (LLMs) through a /setgo skill that enables natural-language execution of the full assess–enrich–publish loop, with user involvement limited to supplying missing metadata values.

Wilkinson, Sean [ORNL] (ORCID:0000000214437479)↗
DOE OSTI2024

Privacy-Preserving Federated Learning for Science: Challenges and Research Directions

This paper discusses the key challenges and future research directions for privacy-preserving federated learning (PPFL), with a focus on its application to large-scale scientific AI models, in particular, foundation models~(FMs). PPFL enables collaborative model training across distributed datasets while preserving privacy-- an important collaborative approach for science. We discuss the need for efficient and scalable algorithms to address the increasing complexity of FMs, particularly when dealing with heterogeneous clients. In addition, we underscore the need for developing advance privacy-preserving techniques, such as differential privacy, to balance privacy and utility in large FMs emphasizing fairness and incentive mechanisms to ensure equitable participation among heterogeneous clients. Finally, we emphasize the need for a robust software stack supporting scalable and secure PPFL deployments across multiple high-performance computing facilities. We envision that PPFL would play a crucial role to advance scientific discovery and enable large-scale, privacy-aware collaborations across science domains.

Kim, Kibaek [Argonne National Laboratory (ANL)]↗
DOE OSTI2020

Better together: Elements of successful scientific software development in a distributed collaborative community

Many scientific disciplines rely on computational methods for data analysis, model generation, and prediction. Implementing these methods is often accomplished by researchers with domain expertise but without formal training in software engineering or computer science. This arrangement has led to underappreciation of sustainability and maintainability of scientific software tools developed in academic environments. Some software tools have avoided this fate, including the scientific library Rosetta. We use this software and its community as a case study to show how modern software development can be accomplished successfully, irrespective of subject area. Rosetta is one of the largest software suites for macromolecular modeling, with 3.1 million lines of code and many state-of-the-art applications. Since the mid 1990s, the software has been developed collaboratively by the RosettaCommons, a community of academics from over 60 institutions worldwide with diverse backgrounds including chemistry, biology, physiology, physics, engineering, mathematics, and computer science. Developing this software suite has provided us with more than two decades of experience in how to effectively develop advanced scientific software in a global community with hundreds of contributors. Here we illustrate the functioning of this development community by addressing technical aspects (like version control, testing, and maintenance), community-building strategies, diversity efforts, software dissemination, and user support. We demonstrate how modern computational research can thrive in a distributed collaborative community. The practices described here are independent of subject area and can be readily adopted by other software development communities

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

MIBiG 3.0: a community-driven effort to annotate experimentally validated biosynthetic gene clusters

Abstract With an ever-increasing amount of (meta)genomic data being deposited in sequence databases, (meta)genome mining for natural product biosynthetic pathways occupies a critical role in the discovery of novel pharmaceutical drugs, crop protection agents and biomaterials. The genes that encode these pathways are often organised into biosynthetic gene clusters (BGCs). In 2015, we defined the Minimum Information about a Biosynthetic Gene cluster (MIBiG): a standardised data format that describes the minimally required information to uniquely characterise a BGC. We simultaneously constructed an accompanying online database of BGCs, which has since been widely used by the community as a reference dataset for BGCs and was expanded to 2021 entries in 2019 (MIBiG 2.0). Here, we describe MIBiG 3.0, a database update comprising large-scale validation and re-annotation of existing entries and 661 new entries. Particular attention was paid to the annotation of compound structures and biological activities, as well as protein domain selectivities. Together, these new features keep the database up-to-date, and will provide new opportunities for the scientific community to use its freely available data, e.g. for the training of new machine learning models to predict sequence-structure-function relationships for diverse natural products. MIBiG 3.0 is accessible online at https://mibig.secondarymetabolites.org/.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

UC LANL Postdoc Entrepreneur Fellowship Cohort Four

Three Laboratory postdoctoral researchers have been selected for this year's cohort of Entrepreneur Fellows. The six-month fellowship program - a joint initiative of the University of California and Los Alamos, now in its fourth year - supports early career scientists explore applications and evaluate market opportunities for their technologies. The Lab selects Entrepreneur Fellows based on the potential for their projects to have both scientific and commercial impact. The fellows receive six months of facilitated training and mentoring, as well as funding to engage in customer discovery, market exploration and technology advancement

36 MATERIALS SCIENCE↗
DOE OSTI2023

Benchmarking for AI for Science

AI has been instrumental for recent developments in a number of domains of the sciences. With several hundred machine learning (ML) algorithms and models, and numerous AI-specific hardware platforms, a common quest for all scientists working on AI for Science is around the selection of machine learning algorithm(s) to solve their domain-specific scientific problems. A number of different initiatives around AI Benchmarking have been set up and have been useful in understanding the benefits of different ML algorithms for different tasks.However, with the majority of these AI Benchmarking initiatives focusing on the conventional notions of benchmarking, where the focus is purely runtime performance (such as training time or inference time), their suitability for benchmarking different ML algorithms for solving scientific problems has been viewed as a performance problem even though both are hardly the same. To make reasonable, explainable, and justifiable advancements in science using AI, it is critical to focus on the merits of these algorithms in handling different domain science problems. In other words, more emphasis must be given on Benchmarking for AI for Science than AI Benchmarking. The vision of the former is not only to assess the performance of ML algorithms, but also to assess, and understand the benefits and merits of different ML algorithms in handling scientific problems. Benchmarking for AI for Science, instead of pure performance focused AI Benchmarking, has several benefits: (i) it has the potential to offer advances in the sciences, much more rapidly than through pure performance-based AI methods, (ii) it will encourage the community to focus on developing better domain-specific AI techniques, particularly given the provision for being able to benchmark different techniques, and (iii) it will encourage hardware manufacturers to focus on developing science-specific hardware subsystems.

Thiyagalingam, Jeyan↗