SEARCH · Engineering Papers

Results for “learning rate”

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 811 records · Page 45

All sources NASA NTRS DOE OSTI

DOE OSTI2023

A hybrid machine-learning approach for analysis of methane hydrate formation dynamics in porous media with synchrotron CT imaging

Fast multi-phase processes in methane hydrate bearing samples pose a challenge for quantitative micro-computed tomography study and experiment steering due to complex tomographic data analysis involving time-consuming segmentation procedures. This is because of the sample's multi-scale structure, which changes over time, low contrast between solid and fluid materials, and the large amount of data acquired during dynamic processes. Here, a hybrid approach is proposed for the automatic segmentation of tomographic data from time-resolved imaging of methane gas-hydrate formation in sandy granular media, which includes a deep-learning 3D U-Net model. To prepare a training dataset for the 3D U-Net, a technique to automate data labeling based on sample-specific information about the mineral matrix immobility and occasional fluid movement in pores is proposed. Automatic segmentation allowed for studying properties of the hydrate growth in pores, as well as dynamic processes such as incremental flow and redistribution of pore brine. Results of the quantitative analysis showed that for typical gas-hydrate stability parameters (100 bar methane pressure, 7°C temperature) the rate of formation is slow (less than 1% per hour), after which the surface area of contact between brine and gas increases, resulting in faster formation (2.5% per hour). Hydrate growth reaches the saturation point after 11 h of the experiment. Finally, the efficacy of the proposed segmentation scheme in on-the-fly automatic data analysis and experiment steering with zooming to regions of interest is demonstrated.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2025

Bond-centric modular design of protein assemblies

Directional interactions that generate regular coordination geometries are a powerful means of guiding molecular and colloidal self-assembly, but implementing such high-level interactions with proteins remains challenging due to their complex shapes and intricate interface properties. Here we describe a modular approach to protein nanomaterial design inspired by the rich chemical diversity that can be generated from the small number of atomic valencies. We design protein building blocks using deep learning-based generative tools, incorporating regular coordination geometries and tailorable bonding interactions that enable the assembly of diverse closed and open architectures guided by simple geometric principles. Experimental characterization confirms the successful formation of more than 20 multicomponent polyhedral protein cages, two-dimensional arrays and three-dimensional protein lattices, with a high (10%–50%) success rate and electron microscopy data closely matching the corresponding design models. Due to modularity, individual building blocks can assemble with different partners to generate distinct regular assemblies, resulting in an economy of parts and enabling the construction of reconfigurable networks for designer nanomaterials.

Biomaterials – proteins↗

DOE OSTI2024

First application of a liquid argon time projection chamber for the search for intranuclear neutron-antineutron transitions and annihilation in 40 Ar using the MicroBooNE detector

We present a novel methodology to search for intranuclear neutron-antineutron transition (n→n̅) followed by n̅-nucleon annihilation within an 40 Ar nucleus, using the MicroBooNE liquid argon time projection chamber (LArTPC) detector. A discovery of n→n̅ transition or a new best limit on the lifetime of this process would either constitute physics beyond the Standard Model or greatly constrain theories of baryogenesis, respectively. The approach presented in this paper makes use of deep learning methods to select n→n̅ events based on their unique features and differentiate them from cosmogenic backgrounds. The achieved signal and background efficiencies are (70.22 ± 6.04)% and (0.0020 ± 0.0003)%, respectively. A demonstration of a search is performed with a data set corresponding to an exposure of 3.32 ×10 26 neutron-years, and where the background rate is constrained through direct measurement, assuming the presence of a negligible signal. With this approach, no excess of events over the background prediction is observed, setting a demonstrative lower bound on the n→n̅ lifetime in 40 Ar of τ m ≳ 1.1×10 26 years, and on the free n→n̅ transition time of τ n→n̅ ≳ 2.6×10 5 s, each at the 90% confidence level. This analysis represents a first-ever proof-of-principle demonstration of the ability to search for this rare process in LArTPCs with high efficiency and low background.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2023

Galaxy Cruise: Deep Insights into Interacting Galaxies in the Local Universe

Abstract We present the first results from GALAXY CRUISE, a community (or citizen) science project based on data from the Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP). The current paradigm of galaxy evolution suggests that galaxies grow hierarchically via mergers, but our observational understanding of the role of mergers is still limited. The data from HSC-SSP are ideally suited to improve our understanding with improved identifications of interacting galaxies thanks to the superb depth and image quality of HSC-SSP. We launched a community science project, GALAXY CRUISE, in 2019 and have collected over two million independent classifications of 20686 galaxies at z < 0.2. We first characterize the accuracy of the participants’ classifications and demonstrate that it surpasses previous studies based on shallower imaging data. We then investigate various aspects of interacting galaxies in detail. We show that there is a clear sign of enhanced activities of super-massive black holes and star formation in interacting galaxies compared to those in isolated galaxies. The enhancement seems particularly strong for galaxies undergoing violent mergers. We also show that the mass growth rate inferred from our results is roughly consistent with the observed evolution of the stellar mass function. The second season of GALAXY CRUISE is currently underway and we conclude with future prospects. We make the morphological classification catalog used in this paper publicly available at the GALAXY CRUISE website, which will be particularly useful for machine-learning applications.

Tanaka, Masayuki↗

DOE OSTI2023

Uncovering New Physics in the Cosmic Microwave Background: Developing Novel Theoretical Models and Machine-Learning-Powered Constraints

The search for evidence of new physics via its signatures in the cosmos is a cornerstone goal of the Office of High Energy Physics in the DOE Office of Science. Indeed, the current concordance cosmological model provides intriguing hints for beyond-the-standard-model (BSM) physics, such as dark matter and dark energy. Recently, a potential breakdown has appeared in this model, which could be initial evidence toward a further important revision in our fundamental theoretical understanding of cosmology. This breakdown is reflected in disagreements between inferences of the current expansion rate of the universe, H 0 (the Hubble constant), based on indirect, cosmological data (e.g., from the early universe) and based on direct, local measurements. Despite significant effort, a compelling new concordance cosmological model has yet to be found; achieving significant progress on this front was the first major focus of the project. Theoretical considerations indicate that if the observational discrepancies are not due to systematic errors, they strongly suggest new physics operating in the redshift range just prior to recombination, when cosmic microwave background (CMB) photons last scattered. Crucially, almost all such models produce unique signatures in the CMB temperature and polarization power spectra, which will be measured with unprecedented precision by ongoing and upcoming experiments, including the DOE-supported CMB-S4 project. However, these subtle hints of new physics must be uncovered from beneath a swath of Galactic and extragalactic foreground contamination. Current CMB analysis methods, although powerful, do not optimally infer the CMB power spectrum in the presence of non-Gaussian foregrounds. There is thus scope for theoretical improvement in this foundational challenge of cosmological inference, which formed the second major focus of the project. The primary objectives of the project were two-fold: (1) to develop new theoretical models in cosmology that can restore concordance amongst the full suite of cosmological data sets, thereby potentially providing evidence of novel BSM physics; (2) to develop new theoretical machinery to enable significant sensitivity improvements in searches for new physics in cosmology, particularly via the CMB power spectrum. The two objectives are intertwined, as the analysis methodology improvements in (2) will enable the tightest possible constraints on the signatures of new physics predicted by the novel scenarios in (1). The theoretical approaches to restore concordance focused on models involving novel scalar field dynamics in the pre-recombination universe (the “early dark energy” scenario and modifications thereof), as well as couplings between this field and other components in the standard cosmological model, such as dark matter. We also studied a model featuring a generalization of the decaying dark matter scenario, in which a sub-component of dark matter converts into dark radiation at late times in cosmic history. While these ideas are mostly driven by phenomenological considerations, this tactic has proven extremely successful in cosmology throughout the past few decades, including in the early history of evidence for dark matter and dark energy. The new theoretical machinery envisioned in (2) is undergirded by developments in signal processing and machine learning, which will enable improvements in CMB power spectrum estimation in the presence of non-Gaussian foreground contaminants. In turn, this will yield optimal sensitivity in searches for new physics in the CMB, by maximizing the cosmological information that is extracted from this observable. The most ambitious outcome of this work would be the construction of a new cosmological model that restores concordance amongst data sets. Although the individual models studied here did not fully achieve that goal, significant progress in narrowing down the model space was made, as described below. Moreover, the outcome of the methodological improvements in (2) will significantly impact a wide range of theoretical cosmology, by enabling the tightest possible constraints on any model that leaves novel signatures in the CMB temperature and polarization power spectra.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2022

ASK: Adversarial Soft k-Nearest Neighbor Attack and Defense

K-Nearest Neighbor (kNN)-based deep learning methods have been applied to many applications due to their simplicity and geometric interpretability. However, the robustness of kNN-based deep classification models has not been thoroughly explored and kNN attack strategies are underdeveloped. In this paper, we first propose an Adversarial Soft kNN (ASK) loss for developing more effective kNN-based deep neural network attack strategies and designing better defense methods against them. Our ASK loss provides a differentiable surrogate of the expected kNN classification error. It is also interpretable as it preserves the mutual information between the perturbed input and the in-class-reference data. We use the ASK loss to design a novel attack method called the ASK-Attack (ASK-Atk), which shows superior attack efficiency and accuracy degradation relative to previous kNN attacks on hidden layers. We then derive an ASK-Defense (ASK-Def) method that optimizes the worst-case ASK training loss. Experiments on CIFAR-10 (ImageNet) show that (i) ASK-Atk achieves ≥13% (≥ 13% ) improvement in attack success rate over previous kNN attacks, and (ii) ASK-Def outperforms the conventional adversarial training method by ≥ 6.9% (≥ 3.5% ) in terms of robustness improvement. Relevant codes are available at https://github.com/wangren09/ASK .

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Assessment of Envelope- and Machine Learning-Based Electrical Fault Type Detection Algorithms for Electrical Distribution Grids

This study introduces envelope- and machine learning (ML)-based electrical fault type detection algorithms for electrical distribution grids, advancing beyond traditional logic-based methods. The proposed detection model involves three stages: anomaly area detection, ML-based fault presence detection, and ML-based fault type detection. Initially, an envelope-based detector identifying the anomaly region was improved to handle noisier power grid signals from meters. The second stage acts as a switch, detecting the presence of a fault among four classes: normal, motor, switching, and fault. Finally, if a fault is detected, the third stage identifies specific fault types. This study explored various feature extraction methods and evaluated different ML algorithms to maximize prediction accuracy. The performance of the proposed algorithms is tested in an emulated software–hardware electrical grid testbed using different sample rate meters/relays, such as SEL735, SEL421, SEL734, SEL700GT, and SEL351S near and far from an inverter-based photovoltaic array farm. The performance outcomes demonstrate the proposed model’s robustness and accuracy under realistic conditions.

24 POWER TRANSMISSION AND DISTRIBUTION↗

DOE OSTI2025

Inference of the Mass Composition of Cosmic Rays with Energies from 10 18.5 to 10 20 eV Using the Pierre Auger Observatory and Deep Learning

We present measurements of the atmospheric depth of the shower maximum X max , inferred for the first time on an event-by-event level using the surface detector of the Pierre Auger Observatory. Using deep learning, we were able to extend measurements of the X max distributions up to energies of 100 EeV ( 10 20 eV ), not yet revealed by current measurements, providing new insights into the mass composition of cosmic rays at extreme energies. Gaining a 10-fold increase in statistics compared to the fluorescence detector data, we find evidence that the rate of change of the average X max with the logarithm of energy features three breaks at 6.5 ± 0.6 ( stat ) ± 1 ( syst ) EeV , 11 ± 2 ( stat ) ± 1 ( syst ) EeV , and 31 ± 5 ( stat ) ± 3 ( syst ) EeV , in the vicinity to the three prominent features (ankle, instep, suppression) of the cosmic-ray flux. The energy evolution of the mean and standard deviation of the measured X max distributions indicates that the mass composition becomes increasingly heavier and purer, thus being incompatible with a large fraction of light nuclei between 50 and 100 EeV. Published by the American Physical Society 2025

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2023

Anomaly Detection in Liquid Sodium Cold Trap Operation with Multisensory Data Fusion Using Long Short-Term Memory Autoencoder

Sodium-cooled fast reactors (SFR), which use high temperature fluid near ambient pressure as coolant, are one of the most promising types of GEN IV reactors. One of the unique challenges of SFR operation is purification of high temperature liquid sodium with a cold trap to prevent corrosion and obstructing small orifices. We have developed a deep learning long short-term memory (LSTM) autoencoder for continuous monitoring of a cold trap and detection of operational anomaly. Transient data were obtained from the Mechanisms Engineering Test Loop (METL) liquid sodium facility at Argonne National Laboratory. The cold trap purification at METL is monitored with 31 variables, which are sensors measuring fluid temperatures, pressures and flow rates, and controller signals. Loss-of-coolant type anomaly in the cold trap operation was generated by temporarily choking one of the blowers, which resulted in temperature and flow rate spikes. The input layer of the autoencoder consisted of all the variables involved in monitoring the cold trap. The LSTM autoencoder was trained on the data corresponding to cold trap startup and normal operation regime, with the loss function calculated as the mean absolute error (MAE). The loss during training was determined to follow log-normal density distribution. During monitoring, we investigated a performance of the LSTM autoencoder for different loss threshold values, set at a progressively increasing number of standard deviations from the mean. The anomaly signal in the data was gradually attenuated, while preserving the noise of the original time series, so that the signal-to-noise ratio (SNR) averaged across all sensors decreased below unity. Results demonstrate detection of anomalies with sensor-averaged SNR < 1.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

DOE OSTI2022

Deep Learning of Dark Energy Spectroscopic Instrument Mock Spectra to Find Damped Lyα Systems

We have updated and applied a convolutional neural network (CNN) machine-learning model to discover and characterize damped Ly α systems (DLAs) based on Dark Energy Spectroscopic Instrument (DESI) mock spectra. We have optimized the training process and constructed a CNN model that yields a DLA classification accuracy above 99% for spectra that have signal-to-noise ratios (S/N) above 5 per pixel. The classification accuracy is the rate of correct classifications. This accuracy remains above 97% for lower S/N ≈1 spectra. This CNN model provides estimations for redshift and H i column density with standard deviations of 0.002 and 0.17 dex for spectra with S/N above 3 pixel -1 . Also, this DLA finder is able to identify overlapping DLAs and sub-DLAs. Further, the impact of different DLA catalogs on the measurement of baryon acoustic oscillations (BAO) is investigated. The cosmological fitting parameter result for BAO has less than 0.61% difference compared to analysis of the mock results with perfect knowledge of DLAs. This difference is lower than the statistical error for the first year estimated from the mock spectra: above 1.7%. We also compared the performances of the CNN and Gaussian Process (GP) models. Our improved CNN model has moderately 14% higher purity and 7% higher completeness than an older version of the GP code, for S/N > 3. Both codes provide good DLA redshift estimates, but the GP produces a better column density estimate by 24% less standard deviation. A credible DLA catalog for the DESI main survey can be provided by combining these two algorithms.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2021

Multi-channel Imager Algorithm (MIA): A novel cloud-top phase classification algorithm

The current Geostationary Operational Environmental Satellites (GOES-16 and 17) cloud-top phase classification algorithm is based primarily on empirical thresholds at multiple wavelengths that have varying absorption capabilities for water and ice. The performance of current GOES-16 cloud-top phase product largely depends on the accuracy of the selection of reflectance ratios. Here this study aims at presenting a novel cloud-top phase classification algorithm (the Multi-channel Imager Algorithm, MIA) that provides a more judicious selection of relationships between channels using a supervised K-mean clustering method on multi-channel Red-Green-Blue images. The K-mean clustering method works analogously to how human eyes separate different colors in a microphysical color rendering set of satellite images, which differentiates water, ice and unclassified thin clouds. For water phase, cloud-top temperature information is used to further distinguish supercooled water. To evaluate the performance of the MIA, an extensive comparison with Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP), Moderate Resolution Imaging Spectroradiometer, and current GOES-16 cloud-top phase products is conducted, using CALIOP as the benchmark. Compared to the current GOES-16 cloud-top phase product, MIA demonstrates a substantial improvement in phase classification, where hit rate increases from 69% to 76% over the Continental United States and 58% to 66% over the full disk domain.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2025

Predictive links between microbial communities and biological oxygen utilization in the Arctic Ocean

Microbial metabolism influences rates of net community production (NCP), exerting a direct biological control on marine oxygen and carbon fluxes. In the Arctic, it is increasingly important to understand and quantify this process, as ecological and oceanographic conditions shift due to changing climate. Here, we describe potential ecological links between pelagic microbial diversity and an NCP precursor, biological oxygen utilization, using machine learning and paired observations of community structure and metabolic activity from a seasonally and spatially variable transect of the Arctic Ocean (2019–2020 MOSAiC Expedition). Community structure was determined using 16S (prokaryotic) and 18S (eukaryotic) rRNA gene amplicon sequencing, and metabolic activity was derived from ΔO 2 /Ar. Using self-organizing maps, we identified clear successional patterns in observed microbial community structure that were seasonally driven in the upper ocean and vertically stratified with depth. Metabolic activity was also stratified, with a primarily net heterotrophic water column (median −1.5% biological oxygen saturation), excepting periodic oxygen supersaturation (maximum: 13.6%) within the mixed layer. Using DNA sequences as predictor variables, we then constructed a random forest regression model that reliably reconstructed biological oxygen concentrations (root mean squared error = 4.14 μmol kg −1 ). Top predictors from this model were from heterotrophic (bacteria) or potentially mixotrophic (dinoflagellate) taxa. These analyses highlight biologically driven diagnostic tools that can be used to expand biogeochemical datasets and improve the microbial perspectives and metabolisms represented in ecological models of net productivity and carbon flux in a changing Arctic Ocean.

Chamberlain, Emelia J. [Univ. of San Diego, San Di↗

DOE OSTI2023

A guide to diagnosing colored resonances at hadron colliders

We present a comprehensive study on how to distinguish the properties of heavy dijet resonances at hadron colliders. A variety of spins, chiral couplings, charges, and QCD color representations are considered. Distinguishing the different color representations is particularly difficult at hadron colliders. To determine the QCD color structure, we consider a third jet radiated in a resonant dijet event. We show that the relative rates of three-jet versus two-jet processes are sensitive to the color representation of the resonance. We also show analytically that the antennae radiation pattern of soft radiation depends on the color structure of dijet events and develops an observable that is sensitive to the antennae patterns. Finally, we exploit a Convolutional Neural Network with Machine Learning techniques to differentiate the radiation patterns from different colored resonances and find encouraging results to discriminate them. We demonstrate our results numerically at a 14 TeV LHC, and the methodology presented here should be applicable to other future hadron colliders.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2022

Efficient, Compact, and Smooth Variable Propulsion Motor (Final Report)

In this project, a new architecture of highly efficient hydraulic motor was developed for the propulsion of off-highway vehicles. The motor uses an adjustable linkage driving a cam to vary the displacement of the piston, resulting in a Variable Displacement Linkage Motor (VDLM). The motor uses low friction rolling element bearings to significantly reduce mechanical friction, especially in the demanding low-speed high-torque conditions experienced by off-highway vehicles. The VDLM has high torque capabilities for its size due to the radial piston packaging and use of a multi-lobe cam. A VDLM is very smooth due to the ability to tune the torque ripple through the design of the cam profile. The project was divided into three periods. During the first period, a dynamic model was constructed of the motor to predict the performance of the motor and the vehicle. During the second period, a single-cylinder learning prototype was designed, built, and tested to validate the models constructed in the first period. In the third period, a multi-cylinder prototype motor was optimized, designed, fabricated, and tested. The motor demonstrated excellent mechanical efficiency (above 92.5% across the range of displacements), but the experimentally measure volumetric efficiency was lower than expected due to higher leakage rates created by the poor tolerance control on the prototype. To validate the dynamic models developed in the first period and better understand design trade-offs. In the third period a multi-cylinder concept demonstration prototype will be designed, fabricated, and tested. The final prototype will be tested on a motor dynamometer and will be utilized in hardware-in-the-loop testing to demonstrate its efficiency and performance impacts on the overall drive train. The experimental results were used in a drive train simulation of a compact track loader operating through a drive cycle. Using the VDLM in a hydrostatic circuit yielded 17.1% reduction in fuel consumption and 36.5% reduction in a series hybrid transmission.

99 GENERAL AND MISCELLANEOUS↗

DOE OSTI2022

FastML Science Benchmarks: Accelerating Real-Time Scientific Edge Machine Learning

Applications of machine learning (ML) are growing by the day for many unique and challenging scientific applications. However, a crucial challenge facing these applications is their need for ultra low-latency and on-detector ML capabilities. Given the slowdown in Moore's law and Dennard scaling, coupled with the rapid advances in scientific instrumentation that is resulting in growing data rates, there is a need for ultra-fast ML at the extreme edge. Fast ML at the edge is essential for reducing and filtering scientific data in real-time to accelerate science experimentation and enable more profound insights. To accelerate real-time scientific edge ML hardware and software solutions, we need well-constrained benchmark tasks with enough specifications to be generically applicable and accessible. These benchmarks can guide the design of future edge ML hardware for scientific applications capable of meeting the nanosecond and microsecond level latency requirements. To this end, we present an initial set of scientific ML benchmarks, covering a variety of ML and embedded system techniques.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2024

Newton-Raphson AC Power Flow Convergence Based on Deep Learning Initialization and Homotopy Continuation

Power flow forms the basis of many power system studies. With the increased penetration of renewable energy, grid planners tend to perform multiple power flow simulations under various operating conditions and not just selected snapshots at peak or light load conditions. Getting a converged AC power flow (ACPF) case remains a significant challenge for grid planners especially in large power grid networks. This paper proposes a two-stage approach to improve Newton-Raphson ACPF convergence and was applied to a 6102 bus Electric Reliability Council of Texas (ERCOT) system. The first stage utilizes a deep learning-based initializer with data re-training. Here a deep neural network (DNN) initializer is developed to provide better initial voltage magnitude and angle guesses to aid in power flow convergence. This is because Newton-Raphson ACPF is quite sensitive to the initial conditions and bad initialization could lead to divergence. The DNN initializer includes a data re-training framework that improves the initializer's performance when faced with limited training data. The DNN initializer successfully solved 3,285 cases out of 3,899 non-converging dispatch and performed better than random forest and DC power flow initialization methods. ACPF cases not solved in this first stage are then passed through a hot-starting algorithm based on homotopy continuation with switched shunt control. The hot-starting algorithm successfully converged 416 cases out of the remaining 614 non-converging ACPF dispatch. In conclusion, the combined two-stage approach achieved a 94.9% success rate, by converging a total of 3,701 cases out of the initial 3,899 unsolved cases.

Deep learning↗

DOE OSTI2025

Improving the fast-charging capability of NbWO-based Li-ion batteries

The discovery of Nb-W-O materials years ago marks the milestone of charging a lithium-ion battery in minutes. Nevertheless, for many applications, charging lithium-ion battery within one minute is urgently demanded, the bottleneck of which largely lies in the lack of fundamental understanding of Li + storage mechanisms in these materials. Herein, by visualizing Li + intercalated into representative Nb 16 W 5 O 55 , we find that the fast-charging nature of such material originates from an interesting rate-dependent lattice relaxation process associated with the Jahn-Teller effect. Furthermore, in situ electron microscopy further reveals a directional, [010]-preferred Li + transport mechanism in Nb 16 W 5 O 55 crystals being the “bottleneck” toward fast charging that deprives the entry of any desolvated Li + through the prevailing non-(010) surfaces. Hence, we propose a machine learning-assisted interface engineering strategy to swiftly collect desolvated Li + and relocate them to (010) surfaces for their fast intercalation. As a result, a capacity of ≈ 116 mAh g −1 (68.5% of the theoretical capacity) at 80 C (45 s) is achieved when coupled with a Li negative electrode.

batteries↗

DOE OSTI2024

Harnessing large language models’ zero-shot and few-shot learning capabilities for regulatory research

Abstract Large language models (LLMs) are sophisticated AI-driven models trained on vast sources of natural language data. They are adept at generating responses that closely mimic human conversational patterns. One of the most notable examples is OpenAI's ChatGPT, which has been extensively used across diverse sectors. Despite their flexibility, a significant challenge arises as most users must transmit their data to the servers of companies operating these models. Utilizing ChatGPT or similar models online may inadvertently expose sensitive information to the risk of data breaches. Therefore, implementing LLMs that are open source and smaller in scale within a secure local network becomes a crucial step for organizations where ensuring data privacy and protection has the highest priority, such as regulatory agencies. As a feasibility evaluation, we implemented a series of open-source LLMs within a regulatory agency’s local network and assessed their performance on specific tasks involving extracting relevant clinical pharmacology information from regulatory drug labels. Our research shows that some models work well in the context of few- or zero-shot learning, achieving performance comparable, or even better than, neural network models that needed thousands of training samples. One of the models was selected to address a real-world issue of finding intrinsic factors that affect drugs' clinical exposure without any training or fine-tuning. In a dataset of over 700 000 sentences, the model showed a 78.5% accuracy rate. Our work pointed to the possibility of implementing open-source LLMs within a secure local network and using these models to perform various natural language processing tasks when large numbers of training examples are unavailable.

Biochemistry & Molecular Biology↗