Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “statistical feature extraction”

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

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

At least 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Bridging the Gap between Cosmological Simulations with Graph Neural Networks and Domain Adaptation

Deep learning models have been shown to outperform methods that rely on summary statistics, like the power spectrum, in extracting information from complex cosmological data sets. However, due to differences in the subgrid physics implementation and numerical approximations across different simulation suites, models trained on data from one cosmological simulation show a drop in performance when tested on another. Similarly, models trained on any of the simulations would also likely experience a drop in performance when applied to observational data. Training on data from two different suites of the CAMELS hydrodynamic cosmological simulations, we examine the generalization capabilities of Domain Adaptive Graph Neural Networks (DA-GNNs). By utilizing GNNs, we capitalize on their capacity to capture structured scale-free cosmological information from galaxy distributions. Moreover, by including unsupervised domain adaptation via Maximum Mean Discrepancy (MMD), we enable our models to extract domain-invariant features. We demonstrate that DA-GNN achieves higher accuracy and robustness on cross dataset tasks (up to 28% better relative error and up to almost an order of magnitude better χ 2 ). Using data visualizations, we show the effects of domain adaptation on proper latent space data alignment. This shows that DA-GNNs are a promising method for extracting domain-independent cosmological information, a vital step toward robust deep learning for real cosmic survey data.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Domain Adaptive Graph Neural Networks for Constraining Cosmological Parameters Across Multiple Data Sets

Deep learning models have been shown to outperform methods that rely on summary statistics, like the power spectrum, in extracting information from complex cosmological data sets. However, due to differences in the subgrid physics implementation and numerical approximations across different simulation suites, models trained on data from one cosmological simulation show a drop in performance when tested on another. Similarly, models trained on any of the simulations would also likely experience a drop in performance when applied to observational data. Training on data from two different suites of the CAMELS hydrodynamic cosmological simulations, we examine the generalization capabilities of Domain Adaptive Graph Neural Networks (DA-GNNs). By utilizing GNNs, we capitalize on their capacity to capture structured scale-free cosmological information from galaxy distributions. Moreover, by including unsupervised domain adaptation via Maximum Mean Discrepancy (MMD), we enable our models to extract domain-invariant features. We demonstrate that DA-GNN achieves higher accuracy and robustness on cross-dataset tasks (up to $28\%$ better relative error and up to almost an order of magnitude better $\chi^2$). Using data visualizations, we show the effects of domain adaptation on proper latent space data alignment. This shows that DA-GNNs are a promising method for extracting domain-independent cosmological information, a vital step toward robust deep learning for real cosmic survey data.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Thresholding Analysis and Feature Extraction from 3D Ground Penetrating Radar Data for Noninvasive Assessment of Peanut Yield

This study explores the efficacy of utilizing a novel ground penetrating radar (GPR) acquisition platform and data analysis methods to quantify peanut yield for breeding selection, agronomic research, and producer management and harvest applications. Sixty plots comprising different peanut market types were scanned with a multichannel, air-launched GPR antenna. Image thresholding analysis was performed on 3D GPR data from four of the channels to extract features that were correlated to peanut yield with the objective of developing a noninvasive high-throughput peanut phenotyping and yield-monitoring methodology. Plot-level GPR data were summarized using mean, standard deviation, sum, and the number of nonzero values (counts) below or above different percentile threshold values. Best results were obtained for data below the percentile threshold for mean, standard deviation and sum. Data both below and above the percentile threshold generated good correlations for count. Correlating individual GPR features to yield generated correlations of up to 39% explained variability, while combining GPR features in multiple linear regression models generated up to 51% explained variability. The correlations increased when regression models were developed separately for each peanut type. This research demonstrates that a systematic search of thresholding range, analysis window size, and data summary statistics is necessary for successful application of this type of analysis. The results also establish that thresholding analysis of GPR data is an appropriate methodology for noninvasive assessment of peanut yield, which could be further developed for high-throughput phenotyping and yield-monitoring, adding a new sensor and new capabilities to the growing set of digital agriculture technologies.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

An adaptive adversarial domain adaptation approach for corn yield prediction

Recently, statistical machine learning and deep learning methods have been widely explored for corn yield prediction. Though successful, machine learning models generated within a specific spatial domain often lose their validity when directly applied to new regions. To address this issue, we designed an unsupervised adaptive domain adversarial neural network (ADANN). Specifically, through domain adversarial training, the ADANN model reduced the impact of domain shift by projecting data from different domains into the same subspace. Also, the ADANN model was designed to be trained in an adaptive way, which guaranteed the model can learn the domain-invariant features and perform accurate yield prediction simultaneously. Informative variables including time-series vegetation indices and sequential weather observations were first collected from multiple data sources and aggregated to the county level. Then, we trained the ADANN model with the extracted features and corresponding reported county-level corn yield from the U.S. Department of Agriculture (USDA). Finally, the trained model was evaluated in four testing years 2016–2019. The U.S. corn belt was used as the study area and counties under study were grouped into two diverse ecological regions. Overall, the experimental results showed that the developed ADANN model had better performance than three other state-of-the-art machine learning models in both local experiments (train and test in the same region) and transfer experiments (train and test in different regions). As the first study using adversarial learning for crop yield prediction, this research demonstrates a novel solution for improving model transferability on crop yield prediction.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Local primordial non-Gaussian bias from time evolution

Primordial non-Gaussianity (PNG) is a signature of fundamental physics in the early Universe that is probed by cosmological observations. Here, it is well known that the local type of PNG generates a strong signal in the two-point function of large-scale structure tracers, such as galaxies. This signal, often termed “scale-dependent bias” is a generic feature of modulation of gravitational structure formation by a large-scale mode. It is less well appreciated that the coefficient controlling this signal, b ϕ , is closely connected to the time evolution of the tracer number density. This correspondence between time evolution and local PNG can be simply explained for a universal tracer whose mass function only depends on peak height and, more generally, for nonuniversal tracers in the separate universe picture, which we validate in simulations. We also describe how to recover the bias of tracers subject to a survey selection function and perform a simple demonstration on simulated galaxies. Since the local PNG amplitude in n-point statistics ($f$ NL ) is largely degenerate with the coefficient b ϕ , this proof of concept study demonstrates that Galaxy survey data can allow for more optimal and robust extraction of local PNG information from upcoming surveys.

Sullivan, James M. [University of California, Berk↗
DOE OSTI2024

Training custom light curve models of SN Ia subpopulations selected according to host galaxy properties

ABSTRACT Type Ia supernova (SN Ia) cosmology analyses include a luminosity step function in their distance standardization process to account for an observed yet unexplained difference in the post-standardization luminosities of SNe Ia originating from different host galaxy populations [e.g. high-mass ($M \gtrsim 10^{10} \, {\rm M}_{\odot }$) versus low-mass galaxies]. We present a novel method for including host-mass correlations in the SALT3 (Spectral Adaptive Light curve Template 3) light curve model used for standardizing SN Ia distances. We split the SALT3 training sample according to host-mass, training independent models for the low- and high-host-mass samples. Our models indicate that there are different average Si ii spectral feature strengths between the two populations, and that the average spectral energy distribution of SNe from low-mass galaxies is bluer than the high-mass counterpart. We then use our trained models to perform an SN cosmology analysis on the 3-yr spectroscopically confirmed Dark Energy Survey SN sample, treating SNe from low- and high-mass host galaxies as separate populations throughout. We find that our mass-split models reduce the Hubble residual scatter in the sample, albeit at a low statistical significance. We do find a reduction in the mass-correlated luminosity step but conclude that this arises from the model-dependent re-definition of the fiducial SN absolute magnitude rather than the models themselves. Our results stress the importance of adopting a standard definition of the SN parameters (x0, x1, c) in order to extract the most value out of the light curve modelling tools that are currently available and to correctly interpret results that are fit with different models.

Taylor, G. (ORCID:0000000157563259)↗
DOE OSTI2023

Application of unsupervised deep learning to image segmentation and in-situ contact angle measurements in a CO 2 -water-rock system

Rock surface wettability is a critical property that regulates multiphase flows in porous media, which can be quantified using the surface contact angle (CA). X-ray micro-computed tomography (μCT) provides an effective approach to in-situ measurements of surface CAs. However, the CA measurement accuracy depends significantly on the quality of CT image segmentation, which is the clustering of CT pixels into separate phases. Inspired by this, we developed a deep learning (DL)-based CA measurement workflow. Motivated by the recent tremendous progress in unsupervised learning techniques and aiming to avoid expensive manual data annotations, an unsupervised DL pipeline for CT image segmentation was proposed and implemented, which includes unsupervised model training and post-processing. The unsupervised model training was driven by a novel loss function constrained with feature similarity and spatial continuity and implemented by iterative forward and backward paths; the former clustered the pixel-wise feature vectors extracted by convolution neural networks, whereas the latter updated the parameters using gradient descent. An over-segmentation strategy was adopted for model training. The post-processing steps based on agglomerative hierarchical clustering (AHC) were implemented to further merge the over-segmented model output to the desired cluster number, which is intended to improve the efficiency of image segmentation. The developed unsupervised DL pipeline was compared with other commonly-used image segmentation methods using pixel-wise and physics-based evaluation metrics on a synthetic raw-image dataset, which had a known ground truth. The unsupervised DL pipeline showed the best performance. Next, the segmented images were input to an automatic CA measurement tool, and the results were validated by comparisons with manual measurements. The CA values from the manual and automatic measurements showed similar distributions and statistical properties. The automatic measurement demonstrated a wider spectrum because of the much larger number of measurement data points. The primary novelty of the unsupervised DL pipeline developed in this study lies in the novel loss function and the over-segmentation strategy associated with AHC post-processing. Finally, the workflow has been proven an efficient tool for pore-scale wettability characterization, which has a wide range of applications in fundamental studies of multiphase flows in natural porous media, which have critical implications to geological carbon sequestration, hydrocarbon energy recovery, and contaminant transport in groundwater.

42 ENGINEERING↗
DOE OSTI2021

FY21 Progress Report: SRNL Analysis of ICCWR LCM and WAMS data for Corrosion and Cracking

The development of algorithms for machine learning and data analysis for the 3013 Surveillance Program is a collaborative effort by the Savannah River National Laboratory (SRNL) and the University of South Carolina (USC). For corrosion detection, Laser Confocal Microscope (LCM) or Wide Area 3D Measurement System (WAMS) data is extracted from large binary files, with software written to convert the data to physical attributes (e.g., height, color and grayscale values; all as functions of a location in a plane projection). A user-friendly Matlab Graphical User Interface (GUI) that reads data from either LCM or WAMS files was developed to integrate input data with software developed for processing and evaluation. The GUI can selectively download binary data, interrogate data attributes, label data, flag significant features, execute Machine Learning (ML) algorithms, output parameters for trained ML algorithms, report ML model accuracy with respect to labeled data, and generate graphical representations for various analyses. Features can be called out by user-specified thresholds, manual labeling or machine learning algorithms when they have been completed. The ability to rapidly label data is important because of the volume of data required for training machine learning algorithms. The GUI has the flexibility to allow addition of improved ML algorithms, methods for data visualization, and statistical computations. Statistical analyses via the GUI include areas of pits within a defined range of pit depths, correlations between Red-Green-Blue (RGB) or grayscale intensity and relative surface height, covariances between values associated with features, and feature histograms. The development of supervised machine learning algorithms, however, has been hindered by a lack of training data. The machine learning algorithms for crack identification are being refined but require improvements to the true positive rate for crack detection. This shortcoming is an artifact of the limited training data currently available, perhaps more so than the structure of the neural networks. At present, the best results are had from a consensus over an ensemble of randomly generated Deep Neural Network (DNN) or Convolutional Neural Network (CNN) algorithms. Although the consensus accuracy method has yielded optimum true positive and true negative rates in excess of 80%, additional validation testing is necessary. In addition to the suite of LCM data that was initially used, and which represents the majority of the work presented in this report, WAMS image data was also reviewed at a preliminary level. The review included a comparison between image resolution and dynamic range for each method. WAMS (ZON file) image data was found to have a pixel pitch of 3.69μm compared to 1 μm for the LCM (vk4 file) data, which implies a lower resolution for the WAMS images. Conversely, the ratio of dynamic range of the WAMS data to the LCM data was approximately 41:20 for height data, suggesting that information from WAMS should more accurately determine the depth of pits. At present, the significance of the greater dynamic range of the WAMS data relative to the LCM data has not yet been evaluated.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2022

Dust and the intrinsic spectral index of quasar variations: hints of finite stress at the innermost stable circular orbit

ABSTRACT We present a study of 9 242 spectroscopically confirmed quasars with multiepoch ugriz photometry from the SDSS Southern Survey. By fitting a separable linear model to each quasar’s spectral variations, we decompose their five-band spectral energy distributions into variable (disc) and non-variable (host galaxy) components. In modelling the disc spectra, we include attenuation by dust on the line of sight through the host galaxy to its nucleus. We consider five commonly used attenuation laws, and find that the best description is by dust similar to that of the Small Magellanic Cloud, inferring a lack of carbonaceous grains from the relatively weak 2175-Å absorption feature. We go on to construct a composite spectrum for the quasar variations spanning 700–8000 Å. By varying the assumed power-law Lν ∝ να spectral slope, we find a best-fitting value α = 0.71 ± 0.02, excluding at high confidence the canonical Lν ∝ ν1/3 prediction for a steady-state accretion disc with a T ∝ r−3/4 temperature profile. The bluer spectral index of the observed quasar variations instead supports the model of Agol & Krolik, and Mummery & Balbus, in which a steeper temperature profile, T ∝ r−7/8, develops as a result of finite magnetically induced stress at the innermost stable circular orbit extracting energy and angular momentum from the black hole spin.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Comparison of Results from Recent NNSA and CEA Measurements of the 239 Pu(n, f) Prompt Fission Neutron Spectrum

The National Nuclear Security Administration (NNSA)/DP French Alternative Energies and Atomic Energy Commission (CEA)/DAM agreement on cooperation on fundamental science is a U.S.-French collaborative effort to combine intellectual and experimental resources and further the relevant nuclear science. Recently, both the NNSA and CEA experimental teams performed high-statistics measurements of the 239 Pu(n, f) prompt fission neutron spectrum (PFNS) at the Los Alamos Neutron Science Center, both of which were recently published in the journal Physical Review C. These separate measurements used the same experimental area and a common neutron detector array, but differ in many aspects, including background assessments, data acquisition systems and philosophies, fission detectors, and PFNS extraction techniques. Hence, some aspects of the experimental methods and associated uncertainties are highly correlated while others are independent. The results from both measurements broke new ground for PFNS measurements given their higher accuracy and more detailed study of corrections necessary for the measured quantity compared to existing literature measurements, and both will significantly impact PFNS nuclear data evaluations for the foreseeable future. Here, the focus of this work is to document a comparison of the results from these distinct measurements in terms of the acquired data, the PFNS results, and the measured average PFNS energies. While systematic differences between the PFNS results are present on the 1–3% level, the acquired data relative to each respective measurement at low incident neutron energies are in remarkable agreement, as are the conclusions regarding the magnitude and position of features in the PFNS relating to second-chance fission, third-chance fission, and pre-equilibrium neutron emission.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2022

Deep Neural Network Algorithm for CMC Microstructure Characterization and Variability Quantification

Microstructure characterization and variability quantification are crucial for understanding ceramic matrix composites (CMCs) mechanical behavior and deformation mechanisms across length scales. Traditionally, analyses of the micrographs obtained from microscopy are labor-intensive. However, with the vast improvement in computer vision (CV) and deep learning (DL), an automated algorithm can be designed to extract essential microstructure variability from micrographs which can then be used to construct a statistically representative volume element (SRVE). The DL-based algorithm spans the taxonomy of microstructure analyses, including semantic segmentation of microstructure constituents, secondary phases, matrix/fiber interface, and defects, and quantifying the microstructure variability in terms of probability distributions. In this work, C/SiNC and SiC/SiNC CMCs microstructures are semantically segmented through a deep convolutional neural network, followed by variability quantification through the implementation of a fully connected regression layer, hence forming a deep regression network. The deep regression network operates in a feedforward regime, in which the neuron output signal traverses through the network in a unidirectional manner. The weight tensor associated with each layer is updated through a backpropagation stochastic gradient descent approach. The input gray-scale image obtained through in-house scanning electron microscope and confocal microscope micrographs is augmented through affine transformations to increase the training set size, which is then processed through four strided convolutional layers. This compresses the image resolution by half at each layer while increasing the image depth by applying different filters (image encoding). The class activation maps (CAMs) corresponding to the applied filters highlight the key architectural features and assist with the semantic segmentation of the microstructure.

Hamza, Mohamed H.↗
DOE OSTI2023

3P Program: Phenotyping X Prediction = Productivity (Final Scientific/Technical Report)

The goal of the 3P Program was to establish integrated, real-time phenotyping and to analyze above- and below-ground plant architecture and total carbon partitioning and allocation to predict heterosis and develop superior crop hybrids by fully leveraging the Sorghum gene pool. There were two overarching themes: 1) the development of a new crop improvement approach utilizing advances in high-throughput phenotyping (HTP), computing, and genomics for public dissemination and 2) leveraging this platform for sorghum crop improvement and commercialization. The Clemson team worked on creating genomic resources and using both statistical learning and high-throughput phenotyping in genomics-assisted breeding. Research was broadly interested in the genetics of carbon partitioning, with the aim of improving crop performance and achieving sustainability. The technology and resources created can be readily found in the public domain and serve to advance scientific understanding of crop genomics and breeding. Genomic prediction was able to identify top crosses to be made, and a hybrid prediction pipeline is in place to drive year-over-year genetic gain. Roots have long been ignored by plant breeders and agronomists, not because they are unimportant but because they are hard to measure. This is an untapped white space of potential insight and innovation. To address this, Hi Fidelity Genetics developed the RootTracker to measure roots in the field on a continuous basis. A database system called RootTracker Tracker was developed to handle data coming from the RootTrackers. In using this device, valuable data was observed for plant breeding, hydrochemical development, and other agricultural biology applications. Carnegie Mellon’s goal was developing new techniques to generate high-resolution 3D models of plants from data collected in the field. The idea was that more useful and more informative phenotypes could be extracted by resolving small features, such as seeds and flowers, and that by modeling in 3D, the spatial structure of plants could be examined. To achieve this, multiple images collected by a new small format structured light stereo imager were fused together. A sorghum panicle modeling pipeline was developed to allow the collection and processing of data. Carolina Seed Systems is an agricultural technology company focused on decarbonizing the agricultural system. Their technology pipeline serves to drive fundamental progress towards creation and distribution of carbon negative crops. The genomic and the engineering technology developed through the 3P Program was leveraged to deliver both value and sustainability from the grower to the consumer. Promising sorghum hybrids were scaled up and commercialized. The overall goal of our research was to integrate, create, and deploy genetic and engineering concepts and technologies to enhance crop productivity in a sustainable fashion. The combination of public and private partners allowed the basic research and hypothesis testing to be quickly accelerated for commercial application by the companies yet maintained that the core framework and academic insights remain in the public domain for continued market disruption, competition, and innovation.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

The entropy of galaxy spectra: how much information is encoded?

Abstract The inverse problem of extracting the stellar population content of galaxy spectra is analysed here from a basic standpoint based on information theory. By interpreting spectra as probability distribution functions, we find that galaxy spectra have high entropy, thus leading to a rather low effective information content. The highest variation in entropy is unsurprisingly found in regions that have been well studied for decades with the conventional approach. We target a set of six spectral regions that show the highest variation in entropy – the 4000 Å break being the most informative one. As a test case with real data, we measure the entropy of a set of high-quality spectra from the Sloan Digital Sky Survey, and contrast entropy-based results with the traditional method based on line strengths. The data are classified into star-forming (SF), quiescent (Q), and active galactic nucleus (AGN) galaxies, and show – independently of any physical model – that AGN spectra can be interpreted as a transition between SF and Q galaxies, with SF galaxies featuring a more diverse variation in entropy. The high level of entanglement complicates the determination of population parameters in a robust, unbiased way, and affects traditional methods that compare models with observations, as well as machine learning (especially deep learning) algorithms that rely on the statistical properties of the data to assess the variations among spectra. Entropy provides a new avenue to improve population synthesis models so that they give a more faithful representation of real galaxy spectra.

Ferreras, Ignacio (ORCID:0000000345843127)↗
DOE OSTI2025

A New Measurement of the Neutron Electric Form Factor with the Super Bigbite Spectrometer Apparatus

Protons and neutrons, collectively known as nucleons, make up the nuclei at the core of atoms which form our world. The nucleon has been under intensive study for over 100 years, and yet we still do not fully understand the internal dynamics which govern properties like its spin or its mass-which contributes to almost all of the visible mass in the universe. These dynamics are governed by quantum chromodynamics(QCD), the predictions of which are experimentally tested at high energy accelerator facilities such as Jefferson Lab. TheGEN-II experiment (E12-09-016) is one such experiment. GEN-II is part of the Super Bigbite Spectrometer (SBS) experimental form factor pro gramme taking place in Hall A at Jefferson Lab, which aims to make precision measurements of the nucleon electromagnetic form factors(EMFFs)at record high values of squared four momentum transfer ¿2. EMFFs describe the electric and magnetic moment distributions within the nucleon. They can be measured through elastic electron scattering off the nucleon, and describe the recoil response of the target nucleon at a given energy scale. GEN-II is a double polarized semi-exclusive beam target asymmetry (BTA) experiment, seeking to measure the electric form factor of the neutron,¿¿ ¿,at three new values of squared four-momentum transfer¿2 =2.92,6.74and9.82GeV². The latter two points being at record high ¿2. The form factor is determined through measuring the BTA of quasielliptical scattering of a neutron from a polarized nuclear target. The experiment utilized the CEBAF accelerator to produce longitudinally polarized electrons up to ~85% polarization, which were scattered off neutrons within a novel polarized helium-3 (³He) target. This new polarized ³He target was employed by building on the technology of its precursors which existed in similar preceding experiments. This target was designed to operate at the high luminosities typical of Hall A, and reached a record breaking combination of polarization and beam intensity known as figure of merit, three times larger than those predecessors. The SBS collaboration designed and constructed two brand new high acceptance spectrometers for these experiments, an electron arm named Bigbite (BB) and a hadron arm named Super Bigbite. Both spectrometers featured a large acceptance EM dipole magnet, and complementary detector systems. The electron arm contained gaseous electronmultipliers(GEMs) which were used for high precision tracking of the scattered electrons, a heavy gas cherenkov (GRINCH) which was used for PID between electrons and pions, a plastic i ii scintillator timing hodoscope to provide high resolution timing of the start of events, and a pair of EM calorimeters (BBCal) which provided energy measurements of detected particles, and provided the experimental trigger. The hadron arm also contained a system of GEMs which will be utilized for future SBS experiments, and a hadron calorimeter designed to provide position, timing and energy measurements of the recoiling nucleon. The calibration of all detector subsystems, beam and target data is discussed, with a focus on novel timing calibrations to the hodoscope and hadron calorimeter. An analysis of selecting quasielliptical events and suppressing background contributions from a number of sources which contaminate the final event sample is given. The largest irreducible backgrounds are found to be from misidentified protons, timing accidentals and inelastic events. The physical asymmetry is measured and used to extract a value for the form factor ratio ¿¿ ¿/¿¿ ¿. High precision ¿2 data for ¿¿ ¿ is used to then extract ¿¿ ¿. This work finds at ¿2 = 2.92 GeV2 that ¿¿ ¿ = 0.0129+0.0019 -0.0020. This result is in statistical agreement with existing fits to world data, and predictions from the constituent quark model and Dyson–Schwinger equations, in this region of ¿2.

Penman, Gary [Univ. of Glasgow, Scotland (United K↗
DOE OSTI2026

Dark Energy Survey Year 3 results: Simulation-based 𝑤CDM inference from weak lensing and galaxy clustering maps with deep learning: Analysis design

Data-driven approaches using deep learning are emerging as powerful techniques to extract non-Gaussian information from cosmological large-scale structure. Here, this work presents the first simulation-based inference (SBI) pipeline that combines weak lensing and galaxy clustering maps in a realistic Dark Energy Survey Year 3 (DES Y3) configuration and serves as preparation for a forthcoming analysis of the survey data. We develop a scalable forward model based on the CosmoGridV1 suite of N-body simulations to generate over one million self-consistent mock realizations of DES Y3 at the map level. Leveraging this large dataset, we train deep graph convolutional neural networks on the full survey footprint in spherical geometry to learn low-dimensional features that approximately maximize mutual information with target parameters. These learned compressions enable neural density estimation of the implicit likelihood via normalizing flows in a ten-dimensional parameter space spanning cosmological 𝑤CDM, intrinsic alignment, and linear galaxy bias parameters, while marginalizing over baryonic, photometric redshift, and shear bias nuisances. To ensure robustness, we extensively validate our inference pipeline using synthetic observations derived from both systematic contaminations in our forward model and independent Buzzard galaxy catalogs. Our forecasts yield significant improvements in cosmological parameter constraints, achieving 2−3× higher figures of merit in the 𝛺 𝑚 − 𝑆 8 plane relative to our implementation of baseline two-point statistics and effectively breaking parameter degeneracies through probe combination. These results demonstrate the potential of SBI analyses powered by deep learning for upcoming Stage-IV wide-field imaging surveys.

Thomsen, A. [Zurich, ETH] (ORCID:0000000203099021)↗
DOE OSTI2025

SigTime: Learning and Visually Explaining Time Series Signatures

Understanding and distinguishing temporal patterns in time series data is essential for scientific discovery and decision-making. For example, in biomedical research, uncovering meaningful patterns in physiological signals can improve diagnosis, risk assessment, and patient outcomes. However, existing methods for time series pattern discovery face major challenges, including high computational complexity, limited interpretability, and difficulty in capturing meaningful temporal structures. Here, to address these gaps, we introduce a novel learning framework that jointly trains two Transformer models using complementary time series representations: shapelet-based representations to capture localized temporal structures and traditional feature engineering to encode statistical properties. The learned shapelets serve as interpretable signatures that differentiate time series across classification labels. Additionally, we develop a visual analytics system—SigTime—with coordinated views to facilitate exploration of time series signatures from multiple perspectives, aiding in useful insights generation. We quantitatively evaluate our learning framework on eight publicly available datasets and one proprietary clinical dataset. Additionally, we demonstrate the effectiveness of our system through two usage scenarios along with the domain experts: one involving public ECG data and the other focused on preterm labor analysis.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Model-Free Measurement of Local Entropy Production and Extractable Work in Active Matter

Time-reversal symmetry breaking and entropy production are universal features of nonequilibrium phenomena. Despite its importance in the physics of active and living systems, the entropy production of systems with many degrees of freedom has remained of little practical significance because the high dimensionality of their state space makes it difficult to measure. Here, in this work, we introduce a local measure of entropy production and a numerical protocol to estimate it. We establish a connection between the entropy production and extractability of work in a given region of the system and show how this quantity depends crucially on the degrees of freedom being tracked. We validate our approach in theory, simulation, and experiments by considering systems of active Brownian particles undergoing motility-induced phase separation, as well as active Brownian particles and E.coli in a rectifying device in which the time-reversal asymmetry of the particle dynamics couples to spatial asymmetry to reveal its effects on a macroscopic scale.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2025

A hybrid neural architecture: Online attosecond x-ray characterization

The emergence of high-repetition-rate x-ray free-electron lasers (XFELs), such as SLAC’s LCLS-II, serves as our canonical example for autonomous controls that necessitate high-throughput diagnostics paired with streaming computational pipelines capable of single-shot analysis with extremely low latency. We present the deterministic characterization with an integrated parallelizable hybrid resolver architecture, a hybrid machine learning framework designed for fast, accurate analysis of XFEL diagnostics using angular streaking-based sinogram images. This architecture integrates convolutional neural networks and bidirectional long short-term memory models to denoise input, identify x-ray sub-spike features, and extract sub-spike relative delays with sub-30 attosecond temporal resolution. Deployed on low-latency hardware, it achieves over 10 kHz throughput with 168.3 μs inference latency, indicating scalability to 14 kHz with field-programmable gate array integration. By transforming regression tasks into classification problems and leveraging optimized error encoding, we achieve high precision with low-latency performance that is critical for real-time streaming event selection and experimental control feedback signals. This represents a key development in real-time control pipelines for next-generation autonomous science, generally, and high repetition-rate x-ray experiments in particular.

Accelerator Physics (physics.acc-ph)↗