SEARCH · Engineering Papers

Results for “Traditional Machine Learning”

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

SourceResearch connectionFrom yearThrough year

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

At least 649 records · Page 36

All sources NASA NTRS DOE OSTI

DOE OSTI2020

Critical Infrastructure Classification via CNN-based Modeling and Image Analysis

With recent advances in the fields of satellite imagery and machine learning we now have the ability to develop explainable deep learning models that enhance critical infrastructure analysis. Funded through Idaho National Laboratory’s (INL) Laboratory Directed Research and Development (LDRD) office we are in the process of developing a deep learning model capable of identifying critical infrastructure facilities and embedded features within those facilities. Utilizing current limit of practice techniques in the machine learning areas of explainability and transfer learning our model, once complete, will have the capacity to be used on multiple different imagery data sets and produce results that not only classify critical infrastructure facilities, but also explain why a critical infrastructure facility was classified as a certain type of facility. These advancements eliminate the ‘black box’ approach deep learning models have had in the past, where a user will have to trust the conclusion of a model without understanding what reasoning when into the model’s classification process. They also expand a model’s usefulness, traditionally a deep learning model will have to use the same data set it was originally trained on. Given the long training times of deep learning models this is impractical in a number of scenarios. By utilizing transfer learning advancements, we are eliminating the need to train our model on the same data set it is then run on to classify critical infrastructure facilities. We are also enabling the analysis and classification of data sets that are potentially too small to be divided into a training and testing data set. Once completed this model can provide a foundation to enhanced critical infrastructure analysis, dependency analysis, and potential disaster relief efforts.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS2022

A Framework for Deep Learning Emulation of Numerical Models With a Case Study in Satellite Remote Sensing

Numerical models based on physics represent the state of the art in Earth system modeling and comprise our best tools for generating insights and predictions. Despite rapid growth in computational power, the perceived need for higher model resolutions overwhelms the latest generation computers, reducing the ability of modelers to generate simulations for understanding parameter sensitivities and characterizing variability and uncertainty. Thus, surrogate models are often developed to capture the essential attributes of the full-blown numerical models. Recent successes of machine learning methods, especially deep learning (DL), across many disciplines offer the possibility that complex nonlinear connectionist representations may be able to capture the underlying complex structures and nonlinear processes in Earth systems. A difficult test for DL-based emulation, which refers to function approximation of numerical models, is to understand whether they can be comparable to traditional forms of surrogate models in terms of computational efficiency while simultaneously reproducing model results in a credible manner. A DL emulation that passes this test may be expected to perform even better than simple models with respect to capturing complex processes and spatiotemporal dependencies. Here, we examine, with a case study in satellite-based remote sensing, the hypothesis that DL approaches can credibly represent the simulations from a surrogate model with comparable computational efficiency. Our results are encouraging in that the DL emulation reproduces the results with acceptable accuracy and often even faster performance. We discuss the broader implications of our results in light of the pace of improvements in high-performance implementations of DL and the growing desire for higher resolution simulations in the Earth sciences.

Bayesian Deep Learning↗

DOE OSTI2022

J-PLUS: Support vector regression to measure stellar parameters

Stellar parameters are among the most important characteristics in studies of stars which, in traditional methods, are based on atmosphere models. However, time, cost, and brightness limits restrain the efficiency of spectral observations. The Javalambre Photometric Local Universe Survey (J-PLUS) is an observational campaign that aims to obtain photometry in 12 bands. Owing to its characteristics, J-PLUS data have become a valuable resource for studies of stars. Machine learning provides powerful tools for efficiently analyzing large data sets, such as the one from J-PLUS, and enables us to expand the research domain to stellar parameters. The main goal of this study is to construct a support vector regression (SVR) algorithm to estimate stellar parameters of the stars in the first data release of the J-PLUS observational campaign. The training data for the parameter's regressions are featured with 12-waveband photometry from J-PLUS and are crossidentified with spectrum-based catalogs. These catalogs are from the Large Sky Area Multi-Object Fiber Spectroscopic Telescope, the Apache Point Observatory Galactic Evolution Experiment, and the Sloan Extension for Galactic Understanding and Exploration. We then label them with the stellar effective temperature, the surface gravity, and the metallicity. Ten percent of the sample is held out to apply a blind test. We develop a new method, a multi-model approach, in order to fully take into account, the uncertainties of both the magnitudes and the stellar parameters. The method utilizes more than 200 models to apply the uncertainty analysis. We present a catalog of 2 493 424 stars with the root mean square error of 160 K in the effective temperature regression, 0.35 in the surface gravity regression, and 0.25 in the metallicity regression. We also discuss the advantages of this multi-model approach and compare it to other machine-learning methods.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2024

Discovering invariant spatial features in electron energy loss spectroscopy images on the mesoscopic and atomic levels

Over the last two decades, Electron Energy Loss Spectroscopy (EELS) imaging with a scanning transmission electron microscope has emerged as a technique of choice for visualizing complex chemical, electronic, plasmonic, and phononic phenomena in complex materials and structures. The availability of the EELS data necessitates the development of methods to analyze multidimensional data sets with complex spatial and energy structures. Traditionally, the analysis of these data sets has been based on analysis of individual spectra, one at a time, whereas the spatial structure and correlations between individual spatial pixels containing the relevant information of the physics of underpinning processes have generally been ignored and analyzed only via the visualization as 2D maps. Here, we develop a machine learning-based approach and workflows for the analysis of spatial structures in 3D EELS data sets using a combination of dimensionality reduction and multichannel rotationally invariant variational autoencoders. This approach is illustrated for the analysis of both the plasmonic phenomena in a system of nanowires and in the core excitations in functional oxides using low loss and core-loss EELS, respectively. The code developed in this manuscript is open sourced and freely available and provided as a Jupyter notebook for the interested reader.

36 MATERIALS SCIENCE↗

DOE OSTI2023

A 3D Implementation of Convolutional Neural Network for Fast Inference

Low latency inference has many applications in edge machine learning. In this paper, we present a run-time configurable convolutional neural network (CNN) inference ASIC design for low-latency edge machine learning. By implementing a 5-stage pipelined CNN inference model in a 3D ASIC technology, we demonstrate that the model distributed on two dies utilizing face-to-face (F2F) 3D integration achieves superior performance. Our experimental results show that the design based on 3D integration achieves 43% better energy-delay product when compared to the traditional 2D technology.

Miniskar, Narasinga Rao↗

DOE OSTI2024

CORN (Crop Optimization Realized through Neuralnets)

Traditional models of predicting plant traits are limited because they often rely on linear assumptions that do not fully capture the complexity of biological interactions and DNA-based markers which are static across environments. This project generated a set of RNA-based data from large multi-environment field trials and combined it with advanced machine learning techniques to account for these complex interactions and improve the accuracy of predictions.

59 BASIC BIOLOGICAL SCIENCES↗

NASA NTRS2017

Weather Impact on Airport Arrival Meter Fix Throughput

Time-based flow management provides arrival aircraft schedules based on arrival airport conditions, airport capacity, required spacing, and weather conditions. In order to meet a scheduled time at which arrival aircraft can cross an airport arrival meter fix prior to entering the airport terminal airspace, air traffic controllers make regulations on air traffic. Severe weather may create an airport arrival bottleneck if one or more of airport arrival meter fixes are partially or completely blocked by the weather and the arrival demand has not been reduced accordingly. Under these conditions, aircraft are frequently being put in holding patterns until they can be rerouted. A model that predicts the weather impacted meter fix throughput may help air traffic controllers direct arrival flows into the airport more efficiently, minimizing arrival meter fix congestion. This paper presents an analysis of air traffic flows across arrival meter fixes at the Newark Liberty International Airport (EWR). Several scenarios of weather impacted EWR arrival fix flows are described. Furthermore, multiple linear regression and regression tree ensemble learning approaches for translating multiple sector Weather Impacted Traffic Indexes (WITI) to EWR arrival meter fix throughputs are examined. These weather translation models are developed and validated using the EWR arrival flight and weather data for the period of April-September in 2014. This study also compares the performance of the regression tree ensemble with traditional multiple linear regression models for estimating the weather impacted throughputs at each of the EWR arrival meter fixes. For all meter fixes investigated, the results from the regression tree ensemble weather translation models show a stronger correlation between model outputs and observed meter fix throughputs than that produced from multiple linear regression method.

Machine Learning Model.↗

DOE OSTI2022

Learning Coagulation Processes With Combinatorial Neural Networks

Abstract Simulating the evolution of a coagulating aerosol or cloud of droplets in a key problem in atmospheric science. We present a proof of concept for modeling coagulation processes using a novel combinatorial neural network (CombNN) architecture. Using two types of data from a high‐detail particle‐resolved aerosol simulation, we show that CombNN models outperform standard neural networks and are competitive in accuracy with traditional state‐of‐the‐art sectional models. These CombNN models could have application in learning coarse‐grained coagulation models for multi‐species aerosols and for learning coagulation models from observed size‐distribution data.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2025

Development of interatomic potential and effect of ordering on defect properties in CrMnV

Developing materials that can withstand extreme environments, such as high radiation doses and elevated temperatures, is crucial for next-generation particle accelerators, including the 2.4 MW Long-Baseline Neutrino Facility. High-Entropy Alloys have emerged as promising candidates for beam window materials due to their superior mechanical strength, corrosion resistance, and radiation tolerance. In this study, we focus on the Cr–Mn–V alloy system, developing and employing machine-learning interatomic potentials (MLIPs) to investigate the formation of an ordered phase and its influence on defect properties. Using hybrid Monte Carlo-Molecular Dynamics simulations, we observe the formation of a B2-ordered phase at lower temperatures, consistent with Density Functional Theory (DFT) predictions. Ordered structures display a bimodal distribution of migration energies and reduced mean square displacement values, indicating suppressed vacancy diffusion. Our results also show that the migration energy barrier varies based on the atomic species, with Mn and V exhibiting the highest and lowest average barriers, respectively. These findings suggest that atomic ordering inhibits defect mobility, potentially enhancing the radiation resistance of CrMnV alloys. The validated MLIP provides a reliable framework for simulations that are faster than traditional DFT while maintaining the accuracy required to study defect and ordering properties.

36 MATERIALS SCIENCE↗

DOE OSTI2024

PAIReD jet: A multi-pronged resonance tagging strategy across all Lorentz boosts

We propose a new approach of jet-based event reconstruction that aims to optimally exploit correlations between the products of a hadronic multi-pronged decay across all Lorentz boost regimes. The new approach utilizes clustered small-radius jets as seeds to define unconventional jets, referred to as PAIReD jets. The constituents of these jets are subsequently used as inputs to machine learning-based algorithms to identify the flavor content of the jet. We demonstrate that this approach achieves higher efficiencies in the reconstruction of signal events containing heavy-flavor jets compared to other event reconstruction strategies at all Lorentz boost regimes. Classifiers trained on PAIReD jets also have significantly better background rejections compared to those based on traditional event reconstruction approaches using small-radius jets at low Lorentz boost regimes. The combined effect of a higher signal reconstruction efficiency and better classification performance results in a two to four times stronger rejection of light-flavor jets compared to conventional strategies at low Lorentz-boosts, and rejection rates similar to classifiers based on large-radius multi-pronged jets at high Lorentz-boost regimes.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2025

Autonomous Synthesis and Inverse Design of Electrochromic Polymers with High Efficiency and Accuracy

Here, the design and synthesis of functional polymers, aimed at targeted properties through specific structures, have long been challenged by their complex and often nonlinear structure–property relationships. Key processes, including knowledge accumulation for predictive design and experimental refinement and validation, are traditionally labor-insensitive and time-consuming, making it difficult to balance accuracy and efficiency. Here, we introduce an accelerated, autonomous system for the on-demand synthesis of electronic polymers that achieves the desired electrochromic functionality with high accuracy and efficiency. Our approach leverages large language model-assisted data mining, a physics-informed copolymer machine learning model, and an AI-driven autonomous robotic workflow in the Polybot lab. Within 72 h, Polybot autonomously synthesized electrochromic polymers (ECPs) with targeted, previously-unreported color values, including green polymers with specific absorption profiles, precisely fine-tuning copolymer structures with a 5% step size in comonomer composition within a three-monomer system. A publicly accessible ECP informatics database has also been created to foster knowledge exchange.

AI-driven Robotic Lab↗

DOE OSTI2025

High-Resolution Model Intercomparison Project phase 2 (HighResMIP2) towards CMIP7

Abstract. Robust projections and predictions of climate variability and change, particularly at regional scales, rely on the driving processes being represented with fidelity in model simulations. Consequently, the role of enhanced horizontal resolution in improved process representation in all components of the climate system continues to be of great interest. Recent simulations suggest the possibility of significant changes in both large-scale aspects of the ocean and atmospheric circulations and in the regional responses to climate change, as well as improvements in representations of small-scale processes and extremes, when resolution is enhanced. The first phase of the High-Resolution Model Intercomparison Project (HighResMIP1) was successful at producing a baseline multi-model assessment of global simulations with model grid spacings of 25–50 km in the atmosphere and 10–25 km in the ocean, a significant increase when compared to models with standard resolutions on the order of 1° that are typically used as part of the Coupled Model Intercomparison Project (CMIP) experiments. In addition to over 250 peer-reviewed manuscripts using the published HighResMIP1 datasets, the results were widely cited in the Intergovernmental Panel on Climate Change report and were the basis of a variety of derived datasets, including tracked cyclones (both tropical and extratropical), river discharge, storm surge, and impact studies. There were also suggestions from the few ocean eddy-rich coupled simulations that aspects of climate variability and change might be significantly influenced by improved process representation in such models. The compromises that HighResMIP1 made should now be revisited, given the recent major advances in modelling and computing resources. Aspects that will be reconsidered include experimental design and simulation length, complexity, and resolution. In addition, larger ensemble sizes and a wider range of future scenarios would enhance the applicability of HighResMIP. Therefore, we propose the High-Resolution Model Intercomparison Project phase 2 (HighResMIP2) to improve and extend the previous work, to address new science questions, and to further advance our understanding of the role of horizontal resolution (and hence process representation) in state-of-the-art climate simulations. With further increases in high-performance computing resources and modelling advances, along with the ability to take full advantage of these computational resources, an enhanced investigation of the drivers and consequences of variability and change in both large- and synoptic-scale weather and climate is now possible. With the arrival of global cloud-resolving models (currently run for relatively short timescales), there is also an opportunity to improve links between such models and more traditional CMIP models, with HighResMIP providing a bridge to link understanding between these domains. HighResMIP also aims to link to other CMIP projects and international efforts such as the World Climate Research Program lighthouse activities and various digital twin initiatives. It also has the potential to be used as training and validation data for the fast-evolving machine learning climate models.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2025

Machine Learning-Assisted Distribution System Network Reconfiguration Problem

High penetration from volatile renewable energy resources in the grid and the varying nature of loads raise the need for frequent line switching to ensure the efficient operation of electrical distribution networks. Operators must ensure maximum load delivery, reduced losses, and the operation between voltage limits. However, computations to decide the optimal feeder configuration are often computationally expensive and intractable, making it unfavorable for real-time operations. This is mainly due to the existence of binary variables in the network reconfiguration optimization problem. To tackle this issue, we have devised an approach that leverages machine learning techniques to reshape distribution networks featuring multiple substations. This involves predicting the substation responsible for serving each part of the network. Hence, it leaves simple and more tractable Optimal Power Flow problems to be solved. This method can produce accurate results in a significantly faster time, as demonstrated using the IEEE 37-bus distribution feeder. Compared to the traditional optimization-based approaches, a feasible solution is achieved approximately ten times faster for all the tested scenarios.

deep neural networks↗

NASA NTRSDate not supplied

Enhancing Air Traffic Control Planning with Automatic Speech Recognition

The decisions made during the Federal Aviation Administration Air Traffic Control System Command Center's planning teleconferences hold significant sway over the National Airspace System. Held every two hours, these teleconferences convene air traffic managers and stakeholders from across the nation to discuss airspace conditions, weather, and constraints, leading to the formulation and adjustment of traffic management initiatives. Given the critical nature of these decisions, the need for accurate and efficient record-keeping is paramount. In recent years, the application of automatic speech recognition has gained popularity across diverse industries, including aviation. While traditional applications focus on transcribing air traffic control communication, this paper explores a unique application of automatic speech recognition by converting the audio from planning teleconferences into text transcriptions. This innovative approach addresses key challenges in the field, presenting potential benefits for quality assurance, real-time participation, and downstream natural language processing tasks. A notable breakthrough in the machine learning community, namely the transformer neural network architecture, forms the backbone of the proposed solution in this paper. The transformer architecture's role in this research represents a paradigm shift in the efficiency of automatic speech recognition models. By reducing the amount of in-domain training data required, this architecture allows for the fine-tuning of such models like Whisper, originally pretrained on vast English speech datasets. The adaptability of the transformer architecture proves invaluable in capturing the nuances of aviation terminology and specific language used in planning teleconferences. Leveraging the Whisper model as a baseline, our research details the fine-tuning and validation using a dataset comprising 20 hours of meticulously transcribed planning teleconferences. Notably, the baseline pretrained Whisper model exhibited a word error rate of 18.77%. Through the fine-tuning process, the model achieved a substantial improvement, demonstrating an impressive performance with a reduced word error rate of 6.82%. This substantial decrease in WER not only highlights the effectiveness of the transformer architecture but also emphasizes the practical advancements achieved through the application of automatic speech recognition in this specific domain. The utilization of automatic speech recognition in planning teleconferences in this work introduces several novelties. Firstly, the creation of text transcriptions offers a valuable tool for quality assurance and facilitates the efficient review of teleconferences. This is an important aspect of the proposed solution, given the time-sensitive and high-stakes nature of decisions made during these meetings. Furthermore, text-searchable transcriptions provide a streamlined approach for locating and validating critical information, potentially saving hours of manual effort in searching through audio recordings. Moreover, our research identifies a key use case for external facilities and stakeholders. In situations where attendance at the planning teleconference is not feasible, having access to text transcriptions in real-time or shortly after the teleconference ends, proves to be a time-saving and informative resource. This feature enhances collaboration and ensures that stakeholders can stay abreast of important discussions and decisions even in their absence. Despite the efficiency gains facilitated by the transformer architecture in automatic speech recognition technology, it is essential to acknowledge the human factors in data creation. Subject matter experts play a crucial role in accurately transcribing planning teleconferences due to the specificity and complexity of the information discussed. The research dataset, consisting of 20 hours of transcribed planning teleconferences, forms the foundation for fine-tuning and validating the Whisper model. The achieved word error rate of 6.82% demonstrates promising advancements, particularly in recognizing essential aviation terminology within the teleconferences. In conclusion, this paper presents a comprehensive exploration of the application of automatic speech recognition in Air Traffic Control System Command Center planning teleconferences, leveraging the transformer architecture for enhanced efficiency. The novel contributions lie in the improved accessibility of decision-making records, real-time participation opportunities for external stakeholders, and the potential for downstream natural language processing advancements. As the aviation industry continues to evolve, the integration of automatic speech recognition technologies holds the promise of revolutionizing decision-making processes and contributing to the overall safety and efficiency of air traffic management.

ATM↗

DOE OSTI2022

EXFOR-NSR PDF database: a system for nuclear knowledge preservation and data curation

Current needs of nuclear science and technology include complete, well-documented, and easily verifiable nuclear data. The complete data records require supporting nuclear bibliography, presently stored in dedicated libraries, in addition, to actual data. Additionally, experimental nuclear reaction data (EXFOR) and Nuclear Science References (NSR) databases contain compilations based on primary (journals) and secondary (conference proceedings, theses, preprints, etc.) publications, and data received from authors via private communications. The secondary library materials and private communications often represent a bottleneck for nuclear data verification, compilation, evaluation, and dissemination activities. To address this issue, bibliographic materials were scanned into PDF (Portable Document Format) files and uploaded in a relational database. The traditional scope of nuclear databases that includes meta-data and numbers derived from data in specialized formats was broadened to accommodate the large volumes of original nuclear data publications. The complete PDF publication files were stored in a relational database as Binary Large OBjects (BLOB). This unique collection of nuclear data compilations and supporting publications generate many opportunities for machine learning applications. The Web interfaces for authorized and public access to the EXFOR-NSR nuclear publications database were implemented at the U.S. National Nuclear Data Center, https://www.nndc.bnl.gov/ and IAEA Nuclear Data Section, https://www-nds.iaea.org/ . The current system is complementary to major nuclear libraries and narrowly focused on nuclear data compilation and evaluation procedures. The contents of the PDF database, details of implementation, and Web interface are described. New capabilities for data curation, knowledge preservation, worldwide dissemination, and natural language processing (NLP) applications are given.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2024

RuralAI in Tomato Farming: Integrated Sensor System, Distributed Computing, and Hierarchical Federated Learning for Crop Health Monitoring

Precision horticulture is evolving due to scalable sensor deployment and machine learning (ML) integration. These advancements boost the operational efficiency of individual farms, balancing the benefits of analytics with autonomy requirements. However, given concerns that affect wide geographic regions (e.g., climate change), there is a need to apply models that span farms. Federated learning (FL) has emerged as a potential solution. FL enables decentralized ML across different farms without sharing private data. Traditional FL assumes simple two-tier network topologies and, thus, falls short of operating on more complex networks found in real-world agricultural scenarios. Networks vary across crops and farms and encompass various sensor data modes, extending across jurisdictions. New hierarchical FL (HFL) approaches are needed for more efficient and context-sensitive model sharing, accommodating regulations across multiple jurisdictions. Here, we present the RuralAI architecture deployment for tomato crop monitoring, featuring sensor field units for soil, crop, and weather data collection. HFL with personalization is used to offer localized and adaptive insights. Model management, aggregation, and transfers are facilitated via a flexible approach, enabling seamless communication between local devices, edge nodes, and the cloud.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2023

ARMing the Edge: Designing Edge Computing–Capable Machine Learning Algorithms to Target ARM Doppler Lidar Processing

Abstract There is a need for long-term observations of cloud and precipitation fall speeds in validating and improving rainfall forecasts from climate models. To this end, the U.S. Department of Energy Atmospheric Radiation Measurement (ARM) user facility Southern Great Plains (SGP) site at Lamont, Oklahoma, hosts five ARM Doppler lidars that can measure cloud and aerosol properties. In particular, the ARM Doppler lidars record Doppler spectra that contain information about the fall speeds of cloud and precipitation particles. However, due to bandwidth and storage constraints, the Doppler spectra are not routinely stored. This calls for the automation of cloud and rain detection in ARM Doppler lidar data so that the spectral data in clouds can be selectively saved and further analyzed. During the ARMing the Edge field experiment, a Waggle node capable of performing machine learning applications in situ was deployed at the ARM SGP site for this purpose. In this paper, we develop and test four algorithms for the Waggle node to automatically classify ARM Doppler lidar data. We demonstrate that supervised learning using a ResNet50-based classifier will classify 97.6% of the clear-air images and 94.7% of cloudy images correctly, outperforming traditional peak detection methods. We also show that a convolutional autoencoder paired with k -means clustering identifies 10 clusters in the ARM Doppler lidar data. Three clusters correspond to mostly clear conditions with scattered high clouds, and seven others correspond to cloudy conditions with varying cloud-base heights.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2025

Fine-tuning machine-learned particle-flow reconstruction for new detector geometries in future colliders

We demonstrate transfer learning capabilities in a machine-learned algorithm trained for particle-flow reconstruction in high energy particle colliders. This paper presents a cross-detector fine-tuning study, where we initially pretrain the model on a large full simulation dataset from one detector design, and subsequently fine-tune the model on a sample with a different collider and detector design. Specifically, we use the Compact Linear Collider detector (CLICdet) model for the initial training set and demonstrate successful knowledge transfer to the CLIC-like detector (CLD) proposed for the Future Circular Collider in electron-positron mode. We show that with an order of magnitude less samples from the second dataset, we can achieve the same performance as a costly training from scratch, across particle-level and event-level performance metrics, including jet and missing transverse momentum resolution. Furthermore, we find that the fine-tuned model achieves comparable performance to the traditional rule-based particle-flow approach on event-level metrics after training on 100,000 CLD events, whereas a model trained from scratch requires at least 1 million CLD events to achieve similar reconstruction performance. To our knowledge, this represents the first full-simulation cross-detector transfer learning study for particle-flow reconstruction. These findings offer valuable insights towards building large foundation models that can be fine-tuned across different detector designs and geometries, helping to accelerate the development cycle for new detectors and opening the door to rapid detector design and optimization using machine learning.

43 PARTICLE ACCELERATORS↗