SEARCH · Engineering Papers

Results for “AI efficiency”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

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

At least 397 records · Page 22

All sources NASA NTRS DOE OSTI

DOE OSTI2021

Earth System Model Improvement Pipeline via Uncertainty Attribution and Active Learning

Primary focal area: 2 (Predictive Modeling via AI): We develop methods to formally quantify uncertainties in Earth System models for the land-atmosphere coupled system. Science Challenge: Earth system models still have significant biases in historical predictions of the intensity and frequency of water cycling extremes (e.g., droughts and flood events), leading to low confidence in future projections. Uncertainties arise from incomplete understanding of land and atmospheric processes, and insufficient observational constraints on model parameters. Many observations, including those from key DOE investments such as ARM and AmeriFlux, are used to evaluate model performance but have not been used to formally quantify model uncertainty because of the expense of running ESM simulations. An efficient pipeline engaging cutting-edge machine learning (ML) and uncertainty quantification (UQ) methods is needed to improve the predictive understanding of water cycle extremes in the Earth system.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2024

Electron and photon efficiencies in LHC Run 2 with the ATLAS experiment

Precision measurements of electron reconstruction, identification, and isolation efficiencies and photon identification efficiencies are presented. They use the full Run 2 data sample collected by the ATLAS experiment in pp collisions at a centre-of-mass energy of 13 TeV during the years 2015–2018, corresponding to an integrated luminosity of 139 fb -1 . The measured electron identification efficiencies have uncertainties that are around 30%–50% smaller than the previous Run 2 results due to an improved methodology and the inclusion of more data. A better pile-up subtraction method leads to electron isolation efficiencies that are more independent of the amount of pile-up activity. Updated photon identification efficiencies are also presented, using the full Run 2 data. When compared to the previous measurement, a 30%–40% smaller uncertainty is observed on the photon identification efficiencies, thanks to the increased amount of available data.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2024

Barriers to adopting artificial intelligence and machine learning technologies in nuclear power

Artificial intelligence and machine learning (AI/ML) technologies offer unique opportunities to transform nuclear plant operations and power generation. Benefits will be felt not only within existing analog and digital instrumentation and control, but also within work processes, the integration of people with technology and most importantly, the business case. The application of this new technology can help simplify complex problems and produce more effective decision-making, making nuclear power safer, more efficient, and more economically viable in the current energy market. Nonetheless, there are potential barriers to its adoption that must be overcome. The purpose of this paper is to categorize, review, and discuss barriers to AI/ML adoption within the nuclear power industry, with a focus on existing commercial reactors. Unique considerations for advanced reactors are also offered. Here we provide a comprehensive overview of the historical, technical, and business barriers that the industry faces, as well as stakeholder readiness, and end-user acceptance. We underscore the importance of user experience and offer potential solutions in overcoming each barrier. These include provisions for easier plant data access, a friendly regulatory environment, and investment in user trust and explainable AI.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DOE OSTI2026

Measurement of differential t-channel single top (anti)quark production cross-sections at 13 TeV with the ATLAS detector

The production of single top quarks and top antiquarks via the t-channel exchange of a virtual W boson is measured in proton-proton collisions at a centre-of-mass energy of 13 TeV at the Large Hadron Collider. The full Run 2 data sample recorded with the ATLAS detector in the years 2015–2018 is used, corresponding to an integrated luminosity of 140 fb−1. The absolute and normalised production cross-sections are measured differentially as a function of the transverse momentum and absolute rapidity of the top quark and top antiquark. In addition, the ratio of top quark to top antiquark production cross-sections is measured. The measured distributions are compared with next-to-leading-order quantum chromodynamics predictions obtained with different combinations of matrix-element generators, parton-shower programs and proton parton distribution functions, as well as to next-to-next-to-leading-order calculations. Overall, good agreement is observed between the measurements and the theoretical predictions. For most measured distributions, the sensitivity to differences between the predictions is limited by the systematic uncertainties in the measurement. The measured differential distributions are also interpreted in an effective field theory approach to constrain the Wilson-Coefficient CQq3,1$$ {C}_{Qq}^{3,1} $$ associated with a four-quark operator. The interpretation accounts for the effect of the selection efficiency, which is altered significantly by non-zero contributions from CQq3,1$$ {C}_{Qq}^{3,1} $$.

Aad, G↗

DOE OSTI2022

Influence of Long-Range Forces on the Transition States and Dynamics of NaCl Ion-Pair Dissociation in Water

We study NaCl ion-pair dissociation in a dilute aqueous solution using computer simulations both for the full system with long-range Coulomb interactions and for a well-chosen reference system with short-range intermolecular interactions. Analyzing results using concepts from Local Molecular Field (LMF) theory and the recently proposed AI-based analysis tool “State predictive information bottleneck” (SPIB), we show that the system with short-range interactions can accurately reproduce the transition rate for the dissociation process, the dynamics for moving between the underlying metastable states, and the transition state ensemble. Contributions from long-range interactions can be largely neglected for these processes because long-range forces from the direct interionic Coulomb interactions are almost completely canceled (>90%) by those from solvent interactions over the length scale where the transition takes place. Thus, for this important monovalent ion-pair system, short-range forces alone are able to capture detailed consequences of the collective solvent motion, allowing the use of physically suggestive and computationally efficient short-range models for the dissociation event. Here, we believe that the framework here should be applicable to disentangling mechanisms for more complex processes such as multivalent ion disassociation, where previous work has suggested that long-range contributions may be more important.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2022

Streaming readout for next generation electron scattering experiments

Current and future experiments at the high-intensity frontier are expected to produce an enormous amount of data that needs to be collected and stored for offline analysis. Thanks to the continuous progress in computing and networking technology, it is now possible to replace the standard ‘triggered’ data acquisition systems with a new, simplified and outperforming scheme. ‘Streaming readout’ (SRO) DAQ aims to replace the hardware-based trigger with a much more powerful and flexible software-based one, that considers the whole detector information for efficient real-time data tagging and selection. Considering the crucial role of DAQ in an experiment, validation with on-field tests is required to demonstrate SRO performance. In this paper, we report results of the on-beam validation of the Jefferson Lab SRO framework. In this work, we exposed different detectors (PbWO-based electromagnetic calorimeters and a plastic scintillator hodoscope) to the Hall-D electron-positron secondary beam and to the Hall-B production electron beam, with increasingly complex experimental conditions. By comparing the data collected with the SRO system against the traditional DAQ, we demonstrate that the SRO performs as expected. Furthermore, we provide evidence of its superiority in implementing sophisticated AI-supported algorithms for real-time data analysis and reconstruction.

47 OTHER INSTRUMENTATION↗

DOE OSTI2021

Demonstration of the Plant Fuel Reload Process Optimization for an Operating PWR

The United States (U.S.) nuclear industry is facing a strong challenge to maintain regulatory-required levels of safety while ensuring economic competitiveness to stay in business. Safety remains a key parameter for all aspects related to the operation of light water reactor (LWR) nuclear power plants (NPPs) and can be achieved more economically by using a risk-informed ecosystem such as that being developed by the Risk-Informed Systems Analysis (RISA) Pathway under the U.S. Department of Energy (DOE) Light Water Reactor Sustainability (LWRS) Program. The LWRS Program is promoting a wide range of research and development (R&D) activities with the goal to maximize both the safety and economically efficient performance of NPPs through improved scientific understanding, especially given that many plants are considering second license renewal. The RISA Pathway has two main goals: (1) the deployment of methodologies and technologies that enable better representation of safety margins and the factors that contribute to cost and safety; and (2) the development of advanced applications that enable cost-effective plant operation. This report summarizes the research outcomes in FY-2021, which the project progressed from the planning and methodology development phase to the early demonstration phase. The highlights of these activities are: (1) the development of a multi-objective optimization process using Genetic Algorithms (GAs); (2) the development and test of an approach for optimization process acceleration using artificial intelligence (AI) that significantly reduces the computational burden; (3) the demonstration of the fuel reload optimization framework for a generic pressurized water reactor (PWR); and (4) the demonstration of limiting design basis accident (DBA) scenarios for evaluation of the transition from deterministic to risk-informed approach for fuel reload optimization.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

DOE OSTI2026

Predicting Adaptively Chosen Observables in Quantum Systems

Recent advances have demonstrated that 𝒪⁡(log 𝑀) measurements suffice to predict 𝑀 properties of arbitrarily large quantum many-body systems. However, these remarkable findings assume that the properties to be predicted are chosen independently of the data. This assumption can be violated in practice, where scientists adaptively select properties after looking at previous predictions. This work investigates the adaptive setting for three classes of observables: local, Pauli, and bounded-Frobenius-norm observables. We prove that Ω⁡(√𝑀) samples of an arbitrarily large unknown quantum state are necessary to predict expectation values of 𝑀 adaptively chosen local and Pauli observables, where the system size scales exponentially and polynomially in 𝑀, respectively. We also present computationally efficient algorithms that achieve this information-theoretic lower bound. In contrast, for bounded-Frobenius-norm observables, we devise an algorithm requiring only 𝒪⁡(log 𝑀) samples, independent of system size. These results highlight the potential pitfalls of adaptivity in analyzing data from quantum experiments and provide algorithmic tools to safeguard against erroneous predictions in quantum experiments.

Machine learning↗

DOE OSTI2026

Optimizing Optical Searches for Supermassive Black Hole Binaries in Active Galactic Nuclei Light Curves: Fourier versus Bayesian Periodicity Detection

Simulations predict that supermassive black hole binaries (SMBHBs) will exhibit periodic brightness variations that may exceed the stochastic variability intrinsic to active galactic nuclei (AGN). In this paper, we simulate SMBHBs with damped random walk (DRW) AGN variability and an added sinusoidal signal from the orbital motion, and test three methods—a generalized Lomb–Scargle periodogram (GLSP), a nested Bayesian sampler (NBS), and a weighted wavelet z-transform (or WWZ)—to determine which is best at recovering the periodicity. Our simulated light curves follow the properties of the Catalina Real-Time Transient Survey (or CRTS), Legacy Survey of Space and Time (LSST), and Zwicky Transient Facility (ZTF) to best inform current and future SMBHB searches. We map a broad range of parameter space and identify which DRW-only light curves best mimic periodicity and pass each method’s model selection. The NBS performs best at detecting periodicity and filtering out DRW-only light curves. Combined candidate selection with both the NBS and GLSP significantly reduces false-positive rates (FPRs) with marginal impact on true-positive rates (TPRs). With this joint model selection pipeline, we find the lowest FPRs in ZTF-like simulations and the highest detection rates in LSST-like simulations. Using a modified computation of the false-alarm probability with GLSP, we efficiently triage LSST AGN light curves (∼10 7 light curves in ∼10–30 hr) and achieve TPRs and FPRs of ∼40% and ∼0.5%, respectively.

Banaszak, Sebastian M. [Vanderbilt Univ., Nashvill↗

DOE OSTI2025

Public Reference Data for Megawatt-Scale Hydrogen Electrolysis – Simulated Wind

The U.S. Department of Energy and the National Laboratory of the Rockies (NLR) demonstrate hydrogen electrolysis, hydrogen compression and storage, and variable hydrogen fuel cell power production using megawatt-scale equipment at NLR’s Flatirons Campus as part of the Advanced Research on Integrated Energy Systems (ARIES) initiative. This dataset represents part of that effort and is intended for academic, national laboratory, industrial, and other stakeholders to plan, design, and validate models of megawatt-scale hydrogen technologies and diverse energy infrastructure nationwide. These data provide a baseline for how existing hydrogen electrolysis technologies perform when coupled with various energy technologies. Future datasets will demonstrate how existing hydrogen fuel cell technologies can provide controllable, dispatchable, and variable power output for artificial intelligence (AI) data centers and other variable loads. This dataset entry describes hydrogen production using a single, simulated wind turbine. The electrolyzer is a 1.25-MW proton exchange membrane type MC250 system manufactured by Nel Hydrogen . While the unit supports up to 2.5 MW of electrolysis, NLR only has a single 1.25-MW electrolysis stack. For the simulated wind energy profiles, NLR used OpenFAST to simulate a 3.4-MW International Energy Agency (IEA) reference wind turbine. The hour-long wind energy profiles varied over wind turbulence intensity (Class A or Class C) and average wind speed (5, 7, or 9 m/s). To match the power limits of the 1.25-MW electrolyzer and 3.4-MW IEA wind turbine most effectively and to maximize the efficiency of hydrogen production at a given average wind speed, the profiles were sometimes scaled by two times. This means that, in some cases, the experimental setup assumed two 1.25-MW electrolyzers were coupled with the wind turbine, representing a total maximum electrolysis load of 2.5 MW. Finally, NLR experimented with two settings for the electrolyzer power supply minimum and maximum current ramp rates (gain and slew): 200 and 400 amperes per second. The simulated profiles were translated from power (kilowatts) to current (amperes) using a curve fit with calibration data and sent to the electrolyzer power supply at 1-Hz frequency. These datasets report relevant hydrogen balance-of-plant and system data, all captured at 1 Hz, including hydrogen mass production measured with an Emerson Coriolis flow meter. Each .zip file represents a single wind turbine electrolysis experiment and is formatted as follows: {technology}-{average wind speed}-{turbulence class}_{number of 1.25 MW electrolyzers connected}-{electrolyzer ramp rate in amperes/second} For instance, “windIEA3.4-5ms-C_2-400.zip” represents the hour-long experiment using the IEA 3.4-MW turbine, subjected to an average wind speed of 5 m/s and Class C wind turbulence, and connected to two 1.25-MW electrolyzers with the power supply set to a maximum current ramp rate (gain and slew) of 400 A/s. Each .zip folder contains the following files: A .csv file containing raw data. An .xlsx file explaining all the fields in the raw data. A .png plot showing the time series of hydrogen production in kilograms per hour, electrolysis power consumption, and input wind turbine power. An experiment labeled “characterization_200.zip” demonstrates the MC250 electrolyzer steady-state response with 30 minute load steps for a total duration of 5 hours. Finally, a .csv file is provided with all simulated wind experiments combined into one dataset labeled "combined_wind_experiments.csv". NLR also built an AI/machine-learning predictive model based on these datasets. The model ingests the electrolyzer current command in amperes, as well as various pressures and temperatures across the system, and predicts hydrogen output in kilograms per hour. The complete model can be found at https://huggingface.co/NatLabRockies/ptmelt-hydrogen-electrolysis .

08 HYDROGEN↗

DOE OSTI2026

An AI-driven framework for evaluating local and state authorities’ permitting processes

The demand for new energy infrastructure is increasing across the United States, but heterogenous permitting processes and embedded requirements across different local jurisdictions can cause project delays, increase “soft costs,” and hinder developer expansion. This study analyzes the variability in local permitting requirements across the U.S. and develops a quantitative approach to describe their clarity and effectiveness in enabling infrastructure project development. By using an Energy Language Model (ELM), a large language model (LLM) for energy technologies, we systematically gathered permitting information from nearly 300 state-, county-, and city-level documents, creating a structured dataset of requirements and procedures on an unprecedented scale and speed. Our analysis revealed that local (city and county) permitting requirement documents are underrepresented compared to state-level guidance documents, which can impede timely and cost-effective installation of new electric infrastructure. Our validation process showed that the final database has an accuracy of approximately 95%. We, further, created a new quantitative method to score permitting requirements for clarity and efficiency, with electric vehicle supply equipment as an initial use case. The average local permitting document scored a 1.8 out of 5, which we interpret as meaning that half of the requirements developers face when installing electric infrastructure are ambiguous, increasing both cost and time. We also created a “Generalized Permit Process”, highlighting common procedural steps and identifying specific opportunities for municipalities to improve their documentation. This research establishes a systematic and scalable framework for evaluating the complexities of local infrastructure permitting processes by combining LLM-powered data collection and quantitative scoring. The framework enables policymakers and developers to identify and mitigate procedural bottlenecks, with the expectation that these improvements can accelerate application review and approval, reduce project costs, and expedite connection to utility distribution grids. As a foundational approach for streamlining local project development processes, this study’s methods are intended to be extended to a wide range of energy applications.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

DOE OSTI2024

Comparative Study of Large Language Model Architectures on Frontier

Large language models (LLMs) have garnered significant attention in both the AI community and beyond. Among these, the Generative Pre-trained Transformer (GPT) has emerged as the dominant architecture, spawning numerous variants. However, these variants have undergone pre-training under diverse conditions, including variations in input data, data preprocessing, and training methodologies, resulting in a lack of controlled comparative studies. Here we meticulously examine two prominent open-sourced GPT architectures, GPT-NeoX and LLaMA, leveraging the computational power of Frontier, the world’s first Exascale supercomputer. Employing the same materials science text corpus and a comprehensive end-to-end pipeline, we conduct a comparative analysis of their training and downstream performance. Our efforts culminate in achieving state-of-the-art performance on a challenging materials science benchmark. Furthermore, we investigate the computation and energy efficiency, and propose a computationally efficient method for architecture design. To our knowledge, these pre-trained models represent the largest available for materials science. Our findings provide practical guidance for building LLMs on HPC platforms.

Yin, Junqi↗

DOE OSTI2024

Performance and calibration of quark/gluon-jet taggers using 140 fb -1 of pp collisions at $\sqrt{s}$ = 13 TeV with the ATLAS detector

The identification of jets originating from quarks and gluons, often referred to as quark/gluon tagging, plays an important role in various analyses performed at the Large Hadron Collider, as Standard Model measurements and searches for new particles decaying to quarks often rely on suppressing a large gluon-induced background. This paper describes the measurement of the efficiencies of quark/gluon taggers developed within the ATLAS Collaboration, using $\sqrt{s}$ = 13 TeV proton–proton collision data with an integrated luminosity of 140 fb -1 collected by the ATLAS experiment. Two taggers with high performances in rejecting jets from gluon over jets from quarks are studied: one tagger is based on requirements on the number of inner-detector tracks associated with the jet, and the other combines several jet substructure observables using a boosted decision tree. A method is established to determine the quark/gluon fraction in data, by using quark/gluon-enriched subsamples defined by the jet pseudorapidity. Differences in tagging efficiency between data and simulation are provided for jets with transverse momentum between 500 GeV and 2 TeV and for multiple tagger working points.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2025

Fully‐Printed Ion Sensor Arrays for Measuring Agricultural Nitrogen and Potassium Concentrations Using Nernstian and AI Models

Abstract The chemical composition of growing media is a key factor for plant growth, impacting agricultural yield and sustainability. However, there is a lack of affordable chemical sensors for ubiquitous nutrient ion monitoring in agricultural applications. This work investigates using fully printed ion‐sensor arrays to measure the concentrations of nitrate, ammonium, and potassium in mixed‐electrolyte media. Ion sensor arrays composed of nitrate, ammonium, and potassium ion‐selective electrodes and a printed silver‐silver chloride (Ag/AgCl) reference electrode are fabricated and characterized in aqueous solutions in a range of concentrations that encompass what is typical for agricultural growing media (0.01 m m –1 m ). The sensors are also tested in mixed‐electrolyte solutions of NaNO 3 , NH 4 Cl, and KCl of varying concentrations, and the recorded potentials are input into Nernstian and artificial neural network models to compare the prediction accuracy of the models against ground truth. The artificial neural network models demonstrated higher accuracy over the Nernstian model, and the model using only ion‐sensor inputs is 7.5% more accurate than the Nernstian model under the same conditions. By enabling more precise and efficient fertilizer application, these sensor arrays coupled to computational models can help increase crop yields, optimize resource use, and reduce environmental impact.

Goodrich, Payton [University of California Berkele↗

DOE OSTI2026

Intelligent Experiments through Real-Time AI: Fast Data Processing and Autonomous Detector Control for High-Energy Nuclear Experiments

The aim of this project is to develop software and hardware for fast real-time data processing and autonomous detector control and calibration for the sPHENIX and the future EIC experiments. Below summarizes Georgia Tech team efforts in the past year: 1. We developed a real-time clustering algorithm and FPGA-based pipeline architecture for processing fired pixel data from ALPIDE sensors in sPHENIX experiments. Our Columnar Clustering Co-Design introduces a hardware-aware, stream-friendly approach that segments pixel data by column pairs using a Column Pair Clustering (CPC) strategy, followed by Cluster Stitching to merge adjacent subclusters. Implemented in Vitis HLS, the pipeline comprises five stages—read-in, subclustering, stitching, analysis, and write-out—connected by tagged HLS streams with custom end-of-event signaling for robust synchronization. We designed a pipelined dataflow model optimized for throughput, low latency, and minimal buffering, enabling scalable clustering across events of arbitrary size. Our system maintains spatial precision via center-of-mass and shape key extraction and efficiently handles edge cases such as fragmented or nested clusters. Compared against DBSCAN in both software and hardware, our approach demonstrates competitive performance under FPGA constraints. 2. We also conducted a comprehensive algorithm-to-hardware co-design of connected component analysis tailored for sPHENIX experiments, focusing on real-time, low-latency processing using FPGAs and High-Level Synthesis (HLS). Starting from a Python-based particle tracking pipeline, the team translated the core logic—graph traversal via DFS and Union-Find—into an HLS-compatible C++ model, replacing dynamic memory and recursion with static arrays and pipelined control flow. The final design includes a fully streamed and dataflow-compatible Union-Find kernel optimized across five iterations, incorporating loop pipelining, array partitioning, AXI/FIFO interface tuning, and function flattening. Experimental results show up to 14.8× speedup over the CPU baseline, reducing per-graph latency to 1.58 μs and demonstrating strong resource efficiency with only ~7k LUTs and zero BRAM usage. The design maintains functional correctness against the Python reference using a Python-based C-simulation framework and Mean Squared Error metrics. This work validates the potential of HLS-driven FPGA designs for edge-level HEP data acquisition, laying a scalable foundation for future integration with real-time detector pipelines and multi-graph processing systems.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2025

Predictions for the Detectability of Milky Way Satellite Galaxies and Outer-Halo Star Clusters with the Vera C. Rubin Observatory

We predict the sensitivity of the Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST) to faint, resolved Milky Way satellite galaxies and outer-halo star clusters. We characterize the expected sensitivity using simulated LSST data from the LSST Dark Energy Science Collaboration (DESC) Data Challenge 2 (DC2) accessed and analyzed with the Rubin Science Platform as part of the Rubin Early Science Program. We simulate resolved stellar populations of Milky Way satellite galaxies and outer-halo star clusters over a wide range of sizes, luminosities, and heliocentric distances, which are broadly consistent with expectations for the Milky Way satellite system. We inject simulated stars into the DC2 catalog with realistic photometric uncertainties and star/galaxy separation derived from the DC2 data itself. We assess the probability that each simulated system would be detected by LSST using a conventional isochrone matched-filter technique. We find that assuming perfect star/galaxy separation enables the detection of resolved stellar systems with $M_V$ = 0 mag and $r_{1/2}$ = 10 pc with >50% efficiency out to a heliocentric distance of ~250 kpc. Similar detection efficiency is possible with a simple star/galaxy separation criterion based on measured quantities, although the false positive rate is higher due to leakage of background galaxies into the stellar sample. When assuming perfect star/galaxy classification and a model for the galaxy-halo connection fit to current data, we predict that 89 +/- 20 Milky Way satellite galaxies will be detectable with a simple matched-filter algorithm applied to the LSST wide-fast-deep data set. Different assumptions about the performance of star/galaxy classification efficiency can decrease this estimate by ~7%-25%, which emphasizes the importance of high-quality star/galaxy separation for studies of the Milky Way satellite population with LSST.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2026

Algorithm-guided experimentation for autonomous AI systems in self-driving laboratories

This presentation summarizes our work in the PrOMMiS project on benchmarking of data-driven optimization algorithms and their applications in self-driving laboratories. This work supports the broader project goal of accelerating the identification of promising separation methods and operating conditions for critical minerals separation processes. We present a systematic benchmarking study of 42 data-driven optimization algorithms on a broad collection of 502 test problems. The results identify BAM, GLCCLUSTER, and MULTIMIN as the most effective optimization solvers, with BAM showing the highest overall performance and solving more than 80% of the benchmark problems. The study also shows that no single solver consistently outperforms the others across all problem types, indicating that our future laboratory applications may benefit from using a small set of strong solvers rather than relying on a single method. The presentation also illustrates an in-silico chemical reactor case study showing that data-driven optimization methods can guide autonomous experimentation in a self-driving laboratory and identify optimal operating conditions within a small number of experiments. Overall, the results provide a basis for selecting efficient optimization methods and demonstrate the practical use of data-driven optimization in self-driving laboratory workflows.

36 MATERIALS SCIENCE↗

DOE OSTI2021

A Survey: Handling Irregularities in Neural Network Acceleration with FPGAs

In the last decade, Artificial Intelligence (AI) through Deep Neural Networks (DNNs) has penetrated virtually every aspect of science, technology, and business. Many types of DNNs have been and continue to be developed, including Convolutional Neural Networks (CNNs), Recurrent Neural Net- works (RNNs), and Graph Neural Networks (GNNs). The overall problem for all of these Neural Networks (NNs) is that their target applications generally pose stringent constraints on latency and throughput, while also having strict accuracy requirements. There have been many previous efforts in creating hardware to accelerate NNs. The problem designers face is that optimal NN models typically have significant irregularities, making them hardware-unfriendly. In this paper, we first define the problems in NN acceleration by characterizing common irregularities in NN processing into 4 types; then we summarize the existing works that handle the four types of irregularities efficiently using hardware, especially FPGAs; finally, we provide a new vision of next-generation FPGA-based NN acceleration: that the emerging heterogeneity in the next-generation FPGAs is the key to achieving higher performance.

Geng, Tong↗