Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “distributed 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.

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

At least 1,171 records · Page 65
All sourcesNASA NTRSDOE OSTI
NASA NTRSDate not supplied

Uniform Dust Deposition System for Dust Tolerance Studies

Future missions to the Moon will require mechanisms and materials to effectively and reliably operate in the presences of lunar regolith. One of the challenges of developing such technologies is terrestrial testing with lunar dust simulants. Dust has a stochastic nature of depositing and it can be difficult to accurately apply it to surfaces of interest. Deposition density, particles size distribution, and percent coverage are all factors that affect the quality of dust mitigation testing. Therefore, to enable such testing the dust mitigation and seals teams at NASA Glenn Research Center (GRC) have developed a system that can uniformly and repeatably deposit lunar simulants on a range of surfaces. Testing of the Dust Deposition System (DDS) and an associatedsimulant preparation methodology have quantified the levels of uniformity and precision for this system. Furthermore, the deposition system has been paired with automated micrograph capturing and machine learning image analysis to correlate the images of dust on a surface to deposition density (in g/cm2). This correlation alleviates the need for tested samples to be accurately weighed with milligram precision and may be a useful tool for in-situ contamination evaluation on the Moon. This paper covers the design and validation testing for the DDS.

Stephen Gerdts↗
DOE OSTI2025

Quantifying and Zoning Urban Heat Island Effects Using Unsupervised Machine Learning

This work explores the Urban Heat Island (UHI) effects in Maricopa County, Arizona, employing a simulation-based approach that combines large-scale building energy modeling with advanced spatial analysis. Utilizing the Automatic Building Energy Modeling (AutoBEM) software suite, we simulated the energy consumption for approximately 1.35 million buildings based on the Model America version 1.0 (MAv1) dataset. Our methodology incorporated spatial analysis at multiple scales, including individual buildings, clusters of zones determined by K-means clustering, and geographical level evaluation based on Zip codes. The results revealed significant variations in energy consumption and heat emissions across different building types and urban zones. High-emission hotspots identified through clustering pointed to areas most contributing to the UHI effects. Zip code-based area analysis further contextualized these findings, offering an urban context-based perspective on emission distribution and informing potential urban energy policies for mitigating UHI effects.

Chowdhury, Shovan [ORNL]↗
DOE OSTI2025

Predicting cutoff L-shells of solar protons using the GPPSn particle dataset

Solar energetic protons (SEPs) arriving at the Earth trigger severe radiation storms in the near-Earth space, directly impacting space missions operating at various altitudes. Therefore, monitoring SEP events and predicting the penetration depths of solar protons are critical for aerospace sectors. Building on previous efforts, here we demonstrate the feasibility of using proton measurements from the Global Prompt Proton Sensor network (GPPSn), enabled by Los Alamos National Laboratory developed combined X-ray dosimeters aboard GPS satellites, to characterize and predict the penetration of solar protons into the geomagnetic field. The inclined medium-Earth-orbits (MEOs) of the global GPS constellation offer a unique advantage of allowing simultaneous measurements of penetrating solar protons inside both open- and closed-field line regions. Therefore, the L-profiles of ∼10s–100 MeV solar protons and their associated cutoff L-shells can be determined from the GPPSn dataset, using predefined threshold proton flux values rather than traditional flux ratios. After examining a list of SEP event intervals across solar cycles 23, 24 and 25—including the 2024 Mother’s Day superstorm, we showcase how the latest GPPSn proton dataset (release v1.10), reprocessed and calibrated, can not only be used to monitor solar proton distributions inside the dynamic geomagnetic field for individual events, but also to derive a new empirical model linking cutoff L-shells with several key space weather parameters. This newly developed SEPCL-MEO model demonstrates high predictive performance; for example, predictions for > 30 MeV solar protons yield a correlation coefficient of 0.85 and performance efficiency of 0.67 when validated against GPPSn observations. Results from this pilot study underscores the scientific and operational value of the GPPSn dataset, and this dataset—when paired with machine-learning techniques—can play a critical role in observing and predicting the effects of future incoming SEP events, including extreme ones.

58 GEOSCIENCES↗
DOE OSTI2022

Interpreting machine learning prediction of fire emissions and comparison with FireMIP process-based models

Annual burned areas in the United States have increased 2-fold during the past decades. With more large fires resulting in more emissions of fine particulate matter, an accurate prediction of fire emissions is critical for quantifying the impacts of fires on air quality, human health, and climate. This study aims to construct a machine learning (ML) model with game-theory interpretation to predict monthly fire emissions over the contiguous US (CONUS) and to understand the controlling factors of fire emissions. The optimized ML model is used to diagnose the process-based models in the Fire Modeling Intercomparison Project (FireMIP) to inform future development. Results show promising performance for the ML model, Community Land Model (CLM), and Joint UK Land Environment Simulator-Interactive Fire And Emission Algorithm For Natural Environments (JULES-INFERNO) in reproducing the spatial distributions, seasonality, and interannual variability of fire emissions over the CONUS. Regional analysis shows that only the ML model and CLM simulate the realistic interannual variability of fire emissions for most of the subregions (r >0.95 for ML and r =0.14~0.70 for CLM), except for Mediterranean California, where all the models perform poorly (r =0.74 for ML and r <0.30 for the FireMIP models). Regarding seasonality, most models capture the peak emission in July over the western US. However, all models except for the ML model fail to reproduce the bimodal peaks in July and October over Mediterranean California, which may be explained by the smaller wind speeds of the atmospheric forcing data during Santa Ana wind events and limitations in model parameterizations for capturing the effects of Santa Ana winds on fire activity. Furthermore, most models struggle to capture the spring peak in emissions in the southeastern US, probably due to underrepresentation of human effects and the influences of winter dryness on fires in the models. As for extreme events, both the ML model and CLM successfully reproduce the frequency map of extreme emission occurrence but overestimate the number of months with extremely large fire emissions. Comparing the fire PM 2.5 emissions from the ML model with process-based fire models highlights their strengths and uncertainties for regional analysis and prediction and provides useful insights into future directions for model improvements.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRSDate not supplied

Uniform Dust Deposition System for Dust Tolerance Studies

Future missions to the Moon will require mechanisms and materials to effectively and reliably operate in the presences of lunar regolith. One of the challenges of developing such technologies is terrestrial testing with lunar dust simulants. Dust has a stochastic nature of depositing and it can be difficult to accurately apply it to surfaces of interest. Deposition density, particles size distribution, and percent coverage are all factors that affect the quality of dust mitigation testing. Therefore, to enable such testing the dust mitigation and seals teams at NASA Glenn Research Center (GRC) have developed a system that can uniformly and repeatably deposit lunar simulants on a range of surfaces. Testing of the Dust Deposition System (DDS) and an associated simulant preparation methodology have quantified the levels of uniformity and precision for this system. Furthermore, the deposition system has been paired with automated micrograph capturing and machine learning image analysis to correlate the images of dust on a surface to deposition density (in g/cm2). This correlation alleviates the need for tested samples to be accurately weighed with milligram precision and may be a useful tool for in-situ contamination evaluation on the Moon. This paper covers the design and validation testing for the DDS.

Dust↗
NASA NTRSDate not supplied

Novel Observing Strategies (NOS) and Earth System Digital Twins (ESDT) for Disaster Resilience

"Earth systems have now been observed continuously for more than 50 years, not only from space but also from airplanes, balloons and in-situ sensors. With the addition of commercial remote sensing providers, many more Internet-of-Things sensors and new NASA and international observatories being planned, these incredible amounts of data will soon be augmented by even larger amounts of diverse data and therefore will become more and more difficult to access, integrate, understand and utilize. At the same time, because of climate change and its impacts, in order to predict, manage, and mitigate the effects of extreme science events and disasters, the information produced by all of this data needs to be optimized, organized and analyzed in such a way that it can be utilized by traditional as well as many new non-traditional users. This calls for the development of observing systems that are agile, coordinated and responsive to events of interest, and for information systems to be dynamic, interactive and fast. With these objectives in mind, the Advanced Information Systems Technology (AIST) Program has been developing technologies that will allow for the development of Novel Observing Strategies (NOS) in a distributed, coordinated fashion (i.e., similar to an “Internet-of-Earth-Things”), as well as for the development of novel information systems or Earth System Digital Twins (ESDT) that will build a digital replica of the past and current states of the Earth systems, will derive forecasts of future states under nominal assumptions and will also offer the capability to investigate many hypothetical evolution scenarios under varying impact assumptions. Both NOS and ESDT will be facilitated by the unprecedented advances of Artificial Intelligence technologies, especially Machine Learning (ML), for extracting relevant information from these large amounts of data, for fusing and assimilating diverse data together and for running complex models faster. Together NOS and ESDT will help build the capabilities needed to anticipate, predict and mitigate the effects of future disasters."

Earth Science Remote Sensing; Information Systems↗
DOE OSTI2025

LTAU-FF: Loss Trajectory Analysis for Uncertainty in atomistic Force Fields

Model ensembles are effective tools for estimating prediction uncertainty in deep learning atomistic force fields. However, their widespread adoption is hindered by high computational costs and overconfident error estimates. In this work, we address these challenges by leveraging distributions of per-sample errors obtained during training and employing a distance-based similarity search in the model latent space. Our method, which we call LTAU (Loss Trajectory Analysis for Uncertainty), efficiently estimates the full probability distribution function of errors for any test point using the logged training errors, achieving speeds that are 2–3 orders of magnitudes faster than typical ensemble methods and allowing it to be used for tasks where training or evaluating multiple models would be infeasible. We apply LTAU towards estimating parametric uncertainty in atomistic force fields (LTAU-FF), demonstrating that it produces well-calibrated confidence intervals and predicts errors that correlate strongly with the true errors for data near the training domain. Furthermore, we show that the errors predicted by LTAU-FF can be used in practical applications for detecting out-of-domain data, tuning model performance, and predicting failure during simulations. We believe that LTAU will be a valuable tool for uncertainty quantification in atomistic force fields and is a promising method that should be further explored in other domains of machine learning.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Automated Integration of Continental-Scale Observations in Near-Real Time for Simulation and Analysis of Biosphere–Atmosphere Interactions

The National Ecological Observatory Network (NEON) is a continental-scale observatory with sites across the US collecting standardized ecological observations that will operate for multiple decades. To maximize the utility of NEON data, we envision edge computing systems that gather, calibrate, aggregate, and ingest measurements in an integrated fashion. Edge systems will employ machine learning methods to cross-calibrate, gap-fill and provision data in near-real time to the NEON Data Portal and to High Performance Computing (HPC) systems, running ensembles of Earth system models (ESMs) that assimilate the data. For the first time gridded EC data products and response functions promise to offset pervasive observational biases through evaluating, benchmarking, optimizing parameters, and training new machine learning parameterizations within ESMs all at the same model-grid scale. Leveraging open-source software for EC data analysis, we are already building software infrastructure for integration of near-real time data streams into the International Land Model Benchmarking (ILAMB) package for use by the wider research community. We will present a perspective on the design and integration of end-to-end infrastructure for data acquisition, edge computing, HPC simulation, analysis, and validation, where Artificial Intelligence (AI) approaches are used throughout the distributed workflow to improve accuracy and computational performance.

Durden, David J.↗
DOE OSTI2023

Investigation of 3D printed lightweight hybrid composites via theoretical modeling and machine learning

Hybrid composites combine two or more different fillers to achieve multifunctional or advanced material properties, such as lightweight and enhanced mechanical properties. The properties of the composites significantly depend on their microstructures, which can be tailored via advanced 3D printing processes. Understanding the process-structure-property relationships is critical to enable the design and engineering of novel hybrid composites for applications in aerospace, automotive, and protective coatings. Here, for this work, we develop 3D printable and lightweight hybrid composites and leverage the conventional design of experiments, a theoretical hybrid model, and an image-driven machine learning (ML) method to investigate their mechanical behaviors. The hybrid composites are formulated with elastomer matrix, microfillers, and thin-shell particles, enabling a significant degree of design freedom of microstructures with densities and mechanical properties varying up to 70% and 91%, respectively. Our statistical analysis indicates that the 3D printing path direction and the microfibers fraction are dominating process parameters with contribution percentages of 45.3% and 57.7% on the specific stiffness and strength, respectively. A hybrid mechanics model is developed based on a simple Weibull distribution function and classical single-filler models to effectively capture the variations in mechanical properties, however, it overestimates the values due to its statistical constraints and idealization of experimental uncertainty. The image-driven ML model leverages the microscale images directly without losing the structural details, shows more accurate predictions with experimental data, and has 48.6% lower root mean square error than the theoretical model.

3D printing↗
DOE OSTI2023

A baseline for ensemble-based, time-resolved inflow reconstruction for a single turbine using large-eddy simulations and latent diffusion models

We are interested in reconstructing winds flowing through a turbine on a second-by-second basis over a 10 min window. Previously, we developed a machine learning algorithm that takes in a snapshot of wind speed measurements and generates ensembles of three-dimensional wind field estimates. Here, we use these estimates as initial conditions in large-eddy simulations and reconstruct atmospheric and turbine response dynamic quantities in a synthetic field campaign. In doing so, we establish a baseline for model validation that future time-aware data assimilation techniques will be compared to. In turbine-free case studies, ground truth wind speeds consistently fall within our estimated wind speed distribution for the first 100 s after the simulation start. In simulations with turbines, the wind estimates show a small bias of 0.10 m s 1 and good correlation of 0.80 during the first 100 s. During this window, our estimates of the Blade 1 bending moment and generator power typically span the ground truth, with the estimate of the former performing better overall. In summary, this approach shows promise as a stand-alone technique for reconstructing real-world inflow and turbine dynamics in 1-2 min windows and as a foundation for future time-aware data assimilation techniques.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
NASA NTRSDate not supplied

Towards a Marine Stratus Climatology on Drizzle Occurrence from CALIPSO

Marine stratus are a predominant feature of our planet with the annual mean coverage exceeding 20%. They strongly reflect sunlight, yet exert only a modest effect on outgoing infrared radiation, providing a significant net cooling to the Earth’s radiative balance. Their formation is coupled to boundary layer circulations that are driven, in part, by cloud top radiative cooling and evaporative cooling from precipitation in downdrafts. Understanding how these cloud systems evolve as the climate changes is a key question that requires additional information on their lifecycle and microphysical properties to accurately represent their behavior in global circulation models. From a large-scale perspective, insight into the microphysical properties of marine stratus at cloud top can be realized through estimates of the effective radius (Re) of the droplet size distributions derived from MODIS observations. Estimates on the occurrence of rain/drizzle are available from CloudSat. Together these observations indicate that precipitation frequently occurs in clouds with higher cloud top Re. This relationship is consistent with the well documented shift in cloud top droplet size distributions towards fewer, yet larger droplets prior the onset of precipitation. Here we report on a new and complementary set observations from the CALIPSO mission. The approach derives an extinction-to-backscatter ratio (Sc, also known as the cloud lidar ratio) using an established relationship that depends on observations of the lidar attenuated backscatter and volume depolarization ratio within the cloud. Because Sc is strongly and inversely related to Re, a change in the derived Sc from higher to lower values corresponds to a change in the droplet size distribution as seen by MODIS. This change in the lidar signals at cloud top clearly identifies clouds that are capable of precipitation. The presentation provides a brief overview of the approach for deriving Sc and compares CALIOP-derived Sc with observations from other techniques. CALIOP classifications of drizzling clouds, based on the retrieved values Sc, are compared to independent, collocated assessments of drizzle occurrence reported in the standard CloudSat data products. Regional and seasonal comparisons highlight the strengths and weaknesses of the two sensors. A machine learning approach that combines information from both CALIOP and CloudSat showcases possible improvements in the global identification of scenes likely to contain rain-bearing clouds.

CALIPSO↗
DOE OSTI2023

Uncertainty Quantified Machine Learning for Street Level Flooding Predictions in Norfolk, Virginia

Everyday citizens, emergency responders, and critical infrastructure can be dramatically affected by street flooding caused by the increasing tidal levels and major storm events magnified by climate change. Low-level flooding, such as nuisance flooding, continues to increase in frequency, especially in cities like Norfolk, Virginia, which can expect nearly 200 flooding events by 2050 [1]. Recently, machine learning (ML) models have been leveraged to produce real-time predictions based on local weather and geographic conditions. However, ML models are known to produce unusual results when presented with data that varies from their training set. For decision-makers to determine the trustworthiness of the model's predictions, ML models need to quantify their prediction uncertainty. This study applies Deep Quantile Regression (DQR) to a previously published, LSTM-based model for hourly water depth predictions [2], and analyzes its out-of-distribution (OOD) performance.

Goldenberg, Steven↗
DOE OSTI2020

Infrared spectroscopy data- and physics-driven machine learning for characterizing surface microstructure of complex materials

There is a need to characterize complex materials and their dynamics under reaction conditions to accelerate materials design. Adsorbate vibrational excitations are selective to adsorbate/surface interactions and infrared (IR) spectra associated with activating adsorbate vibrational modes are accurate, capture details of most modes, and can be obtained operando. Current interpretation depends on heuristic peak assignments for simple spectra, precluding the possibility of obtaining detailed structural information. Here, we combine data-based approaches with chemistry-dependent problem formulation to develop physics-driven surrogate models that generate synthetic IR spectra from first-principles calculations. Using synthetic IR spectra of carbon monoxide on platinum, we implement multinomial regression via neural network ensembles to learn probability distributions functions (pdfs) that describe adsorption sites and quantify uncertainty. We use these pdfs to infer detailed surface microstructure from experimental spectra and extend this methodology to other systems as a first step towards characterizing complex interfaces and closing the materials gap.

Science & Technology - Other Topics↗
DOE OSTI2026

Short-term electricity load forecasting: Application-driven evaluation of machine learning models across spatial and temporal scales

As we transition towards a decarbonized economy, the integration of variable renewable energy resources and new demands (e.g., electric vehicles, heat pumps) into the electricity grid places unprecedented pressure on grid operators to effectively anticipate and manage peak load. In this context, machine learning algorithms are proving to be indispensable for accurate short-term load forecasting, a crucial task to address these challenges. This study benchmarks 6 machine learning algorithms, including three neural networks and three tree-based algorithms, across various levels of spatial aggregation and time horizons (1, 4, 8, 24, and 48 h). The central contribution of this work is the comparison and analysis of load forecasting models not only based on statistical metrics, but also based on a novel error metric, which evaluates the cost implications of forecast errors for power system stakeholders. Results show that tree-based models outperform neural networks, based on statistical metrics, and yield less skewed error distributions for most spatial scales. However, through the lens of the novel error metric, neural networks are the more competitive choice, especially for forecast horizons that exceed 8 h. The study concludes with actionable recommendations to grid operators and highlights the need for the development of error metrics that link forecasting accuracy to operational costs. To promote transparency and open science, the datasets and Python code are open-sourced via a supplementary repository.

Houben, Nikolaus↗
DOE OSTI2024

Generative machine learning for detector response modeling with a conditional normalizing flow

In this paper, we explore the potential of generative machine learning models as an alternative to the computationally expensive Monte Carlo (MC) simulations commonly used by the Large Hadron Collider (LHC) experiments. Our objective is to develop a generative model capable of efficiently simulating detector responses for specific particle observables, focusing on the correlations between detector responses of different particles in the same event and accommodating asymmetric detector responses. Here, we present a conditional normalizing flow model ($\mathcal{CNF}$) based on a chain of Masked Autoregressive Flows, which effectively incorporates conditional variables and models high-dimensional density distributions. We assess the performance of the $\mathcal{CNF}$ model using a simulated sample of Higgs boson decaying to diphoton events at the LHC. We create reconstruction-level observables using a smearing technique. We show that conditional normalizing flows can accurately model complex detector responses and their correlation. This method can potentially reduce the computational burden associated with generating large numbers of simulated events while ensuring that the generated events meet the requirements for data analyses. We make our code available at https://github.com/allixu/normalizing_flow_for_detector_response

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2021

Exploring the physics of cesium lead halide perovskite quantum dots via Bayesian inference of the photoluminescence spectra in automated experiment

The unique optoelectronic properties of metal halide perovskite quantum dots (QDs) make them promising candidates for applications in light-emitting diodes (LEDs), scintillators, and other photonic devices. The automated micropipetting synthesis platform equipped with an optical reader enables the opportunity for high throughput synthesis and photoluminescent (PL) characterization of metal halide perovskite QDs for the first time. Here, we explore the compositional dependence of the PL behavior and stability of the combinatorial library of cesium lead halide (CsPbX 3 ) perovskites QDs via the automated platform. To study the stability of synthesized QDs in the binary and ternary configurations, we study the time-dependent PL properties using previously developed machine learning analysis. Additionally, to systematically explore the PL behavior in the ternary CsPbX 3 QDs system, we introduce the Bayesian inference framework that allows the probabilistic fit of multiple models to the PL data and establishes both optimal model and model parameter robustly. Furthermore, these behaviors can be used as a control parameter for the navigation of the multidimensional compositional spaces in automated synthesis. Furthermore, this analysis shows the nonuniformity of the PL peak behavior in the ternary CsPbX 3 QDs system. Further, the analysis confirms narrow size distribution and good quality of CsPbBr 3 QDs alloyed with low concentrations of iodide and chloride. We note that Bayesian Inference fit parameters can be further used as a control signal for navigation of the chemical spaces in automated synthesis.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Quantification of the Crack Evolution Process by Extracting Relevant Signal Components from Wave Propagation and Diffusive Transport Front Measurements

Wave propagation and diffusive transport phenomena in a geological rock sample undergoing crack evolution process are expected to interact with the mechanical discontinuities in the medium. The measurements of the signals associated with these phenomena can be used to assess and monitor the crack-driven micromechanical alterations in the rock. Different wave/diffusion phenomena, such as sonic propagation, pressure diffusion, and acoustic emission (AE), are sensitive to different elements of the mechanical discontinuities generated during the evolution of the crack clusters from initiation to coalescence. Sonic propagation, AE, and pressure diffusion monitoring have the potential to map the crack evolution because the transmitter-receiver arrays can be designed, arranged and tuned to (1) achieve maximum recovery of the scattered waveforms and travel times, (2) capture the later arrivals and multiple reflections, and (3) illuminate large rock volume. However, the structural/topological complexities of the mechanical discontinuities, complex distribution of the stress fields, complex mechanical alterations in media, and fluid redistribution in the crack system pose serious challenges for the detection and modeling of the crack evolution process (from here on, we will use the term ‘crack evolution process’ to mean that the crack evolution occurred under shallow crustal conditions). For purposes of accurately accounting such complexities and heterogeneities in the absence of reliable physical laws, simulation methods, and signal processing techniques, my early-career research proposal will develop and apply novel data-driven machine learning methods to: (1) extract signal components relevant to the various phases of crack evolution and (2) generate a 2D visual map of the crack evolution process.

58 GEOSCIENCES↗
DOE OSTI2025

A Data-Agnostic, Continuous Machine Learning Framework for Application in High Energy Physics and Beyond: Phase 1 Final Scientific/Technical Report

This Phase 1 effort has focused on the development of continual learning frameworks for use in machine learning, specifically in the applied context of High Energy Physics (HEP). Machine learning (ML) is a transformative technology by which computers, typically through the use of neural networks, are able to perform tasks with proficiency that rivals or surpasses that of human users. Model Degradation & Catastrophic Forgetting are two undesired phenomena which can occur in ML where the performance of a model degrades when either deployed on novel data streams, or trained on novel data which are sufficiently different than the data the models were initially trained on. A natural example where these sorts of effects can be observed is in the performance of detectors in harsh environments, where the detector signature may change over the lifetime of the detector as it ages and deteriorates — precisely what occurs in the experiments conducted in HEP. Real world HEP data is therefore an excellent test-ground and use-case for Continual Learning paradigms, which are techniques used in ML to counteract these problems. Ensemble learning is one such technique, where multiple smaller models are trained on subsets of the overall data and are ensembled together during inference. The intuition behind this technique is that, although there are shifts in the distributions which govern the incoming data streams, these shifts are not expected to be homogeneous or global. If a sufficient diversity in solutions within the various sub-models has been achieved, then at least one sub-model is expected to retain its performance within the overall ensemble. One further strength of this approach is that the architectures of the various models do not need to be identical, and in fact even different modalities of data can naturally be combined in this way. This work focused on applying ensemble learning techniques to derive results using two main datasets, anomaly detection in HEP data & time-series forecasting in semiconductor manufacturing data. Semiconductor manufacturing involves data with surprising similarity to that of HEP (e.g. wafer maps look very similar to digi-occupancy maps) and Cerium Lab’s prominence within the semiconductor industry makes semiconductor manufacturing a natural opportunity for commercialization of this work. Our efforts have led to two strong results. The first is that we evaluated the proposed ensembling techniques using previously proposed machine learning architectures for use in anomaly detection, namely AutoEncoder based models and their derivatives. We also developed new architectures which have not been evaluated in this context before. In fact, this work marks the first use of Vision Transformers for anomaly detection in HEP. Second, we demonstrated that ensemble learning significantly improves model performance in scenarios prone to degradation, validating its effectiveness across both HEP and semiconductor datasets. These results further support ensemble learning as a powerful strategy for mitigating catastrophic forgetting and maintaining robust performance in evolving data environments.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗