An Uncertainty Quantification Framework for Counter Unmanned Aircraft Systems Using Deep Ensembles.
Abstract not provided.
Sahay, Rajeev↗
SEARCH · Search NASA
Results for “Deep uncertainty”
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.
Reconstruction of fast neutron direction in segmented organic detectors using deep learning
A method for reconstructing the direction of a fast neutron source using a segmented organic scintillator-based detector and deep learning model is proposed and analyzed. Here, the model is based on recurrent neural network, which can be trained by a sequence of data obtained from an event recorded in the detector and suitably pre-processed. The performance of deep learning-based model is compared with the conventional double-scatter detection algorithm in reconstructing the direction of a fast neutron source. With the deep learning model, the uncertainty in source direction of 0.301 rad is achieved with 100 neutron detection events in a segmented cubic organic scintillator detector with a side length of 46 mm. To reconstruct the source direction with the same angular resolution as the double-scatter algorithm, the deep learning method requires 75% fewer events. Application of this method could augment the operation of segmented detectors operated in the neutron scatter camera configuration for applications such as special nuclear material detection.
46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
Near Repository Unsaturated Alluvium Disposal Modeling with Improved Geological Realism
The Disposal Research and Development (R&D) Program of the US Department of Energy (DOE) office of Nuclear Energy (NE-8) Spent Fuel and Waste Science and Technology (SFWST) Campaign is to conduct R&D on disposal of spent nuclear fuel (SNF) and high-level waste (HLW). The goal of the Geologic Disposal Safety Assessment (GDSA) within this project is to develop a disposal system modeling and analysis capability that supports the integrated modeling of coupled processes controlling disposal system performance of deep geologic repositories, including uncertainty. This report describes a specific activity in the Fiscal Year 2024 (FY24) associated with the GDSA Repository Systems Analysis (RSA) work package in collaboration with the GDSA Geologic Modeling work package at Los Alamos National Laboratory (LANL). The overall objective of the GDSA RSA work package is to develop generic deep geologic repository concepts and repository system performance models in crystalline, argillite, salt, and unsaturated alluvium potential host-rock environments, and to simulate and analyze these generic repository concepts and models using GDSA Framework toolkit, and other tools as needed.
12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
Deep learning multiphysics network for imaging CO 2 saturation and estimating uncertainty in geological carbon storage
Multiphysics inversion exploits different types of geophysical data that often complement each other and aims to improve overall imaging resolution and reduce uncertainties in geophysical interpretation. Despite the advantages, traditional multiphysics inversion is challenging because it requires a large amount of computational time and intensive human interactions for preprocessing data and finding trade-off parameters. These issues make it nearly impossible for traditional multiphysics inversion to be applied as a real-time monitoring tool for geological carbon storage. In this paper, we present a deep learning (DL) multiphysics network for imaging CO 2 saturation in real time. The multiphysics network consists of three encoders for analysing seismic, electromagnetic and gravity data and shares one decoder for combining imaging capabilities of the different geophysical data for better predicting CO 2 saturation. The network is trained on pairs of CO 2 label models and multiphysics data so that it can directly image CO 2 saturation. Here we use the bootstrap aggregating method to enhance the imaging accuracy and estimate uncertainties associated with CO 2 saturation images. Using realistic CO 2 label models and multiphysics data derived from the Kimberlina CO 2 storage model, we evaluate the performance of the deep learning multiphysics network and compare its imaging results to those from the deep learning single-physics networks. Our modelling experiments show that the deep learning multiphysics network for seismic, electromagnetic, and gravity data not only improves the imaging accuracy but also reduces uncertainties associated with CO 2 saturation images. Our results also suggest that the deep learning multiphysics network for the non-seismic data (i.e., electromagnetic and gravity) can be used as an effective low-cost monitoring tool in between regular seismic monitoring.
58 GEOSCIENCES↗
Automated segmentation of porous thermal spray material CT scans with predictive uncertainty estimation
Abstract Thermal sprayed metal coatings are used in many industrial applications, and characterizing the structure and performance of these materials is vital to understanding their behavior in the field. X-ray computed tomography (CT) enables volumetric, nondestructive imaging of these materials, but precise segmentation of this grayscale image data into discrete material phases is necessary to calculate quantities of interest related to material structure. In this work, we present a methodology to automate the CT segmentation process as well as quantify uncertainty in segmentations via deep learning. Neural networks (NNs) have been shown to excel at segmentation tasks; however, memory constraints, class imbalance, and lack of sufficient training data often prohibit their deployment in high resolution volumetric domains. Our 3D convolutional NN implementation mitigates these challenges and accurately segments full resolution CT scans of thermal sprayed materials with maps of uncertainty that conservatively bound the predicted geometry. These bounds are propagated through calculations of material properties such as porosity that may provide an understanding of anticipated behavior in the field.
Martinez, Carianne↗
Integrating deep neural networks with full-waveform inversion: Reparameterization, regularization, and uncertainty quantification
Full-waveform inversion (FWI) is an accurate imaging approach for modeling the velocity structure by minimizing the misfit between recorded and predicted seismic waveforms. However, the strong nonlinearity of FWI resulting from fitting oscillatory waveforms can trap the optimization in local minima. We have adopted a neural-network-based full-waveform inversion (NNFWI) method that integrates deep neural networks with FWI by representing the velocity model with a generative neural network. Neural networks can naturally introduce spatial correlations as regularization to the generated velocity model, which suppresses noise in the gradients and mitigates local minima. Furthermore, the velocity model generated by neural networks is input to the same partial differential equation (PDE) solvers used in conventional FWI. The gradients of the neural networks and PDEs are calculated using automatic differentiation, which back propagates gradients through the acoustic PDEs and neural network layers to update the weights of the generative neural network. Experiments on 1D velocity models, the Marmousi model, and the 2004 BP model determine that NNFWI can mitigate local minima, especially for imaging high-contrast features such as salt bodies, and it significantly improves the inversion in the presence of noise. Adding dropout layers to the neural network model also allows analyzing the uncertainty of the inversion results through Monte Carlo dropout. NNFWI opens a new pathway to combine deep learning and FWI for exploiting the characteristics of deep neural networks and the high accuracy of PDE solvers. Because NNFWI does not require extra training data and optimization loops, it provides an attractive and straightforward alternative to conventional FWI.
58 GEOSCIENCES↗
Robust deep learning framework for constitutive relations modeling
Modeling the full-range deformation behaviors of materials under complex loading and materials conditions is a significant challenge for constitutive relations (CRs) modeling. Here, we propose a general encoder-decoder deep learning framework that can model high-dimensional stress-strain data and complex loading histories with robustness and universal capability. The framework employs an encoder to project high-dimensional input information (e.g., loading history, loading conditions, and materials information) to a lower-dimensional hidden space and a decoder to map the hidden representation to the stress of interest. We evaluated various encoder architectures, including gated recurrent unit (GRU), GRU with attention, temporal convolutional network (TCN), and the Transformer encoder, on two complex stress-strain datasets that were designed to include a wide range of complex loading histories and loading conditions. All architectures achieved excellent test results with an root-mean-square error (RMSE) below 1 MPa. Additionally, we analyzed the capability of the different architectures to make predictions on out-of-domain applications, with an uncertainty estimation based on deep ensembles. The proposed approach provides a robust alternative to empirical/semi-empirical models for CRs modeling, offering the potential for more accurate and efficient materials design and optimization.
36 MATERIALS SCIENCE↗
Uncertainty guided online ensemble for non-stationary data streams in fusion science
Machine Learning (ML) is poised to play a pivotal role in the development and operation of next-generation fusion devices. Fusion data shows non-stationary behavior with distribution drifts, resulted by both experimental evolution and machine wear-and-tear. ML models assume stationary distribution and fail to maintain performance when encountered with such non-stationary data streams. Online learning techniques have been leveraged in other domains, however it has been largely unexplored for fusion applications. In this paper, we investigate online learning for continuous adaptation to drifting data streams in the prediction of Toroidal Field (TF) coils deflection at the DIII-D fusion facility. We further address the short-term performance degradation inherent to standard online learning, which arises because ground truth is unavailable at prediction time. To mitigate this issue, we propose an uncertainty-guided online ensemble framework. The method leverages the Deep Gaussian Process Approximation (DGPA) for calibrated uncertainty estimation and uses these uncertainty measures to guide a meta-algorithm that aggregates predictions from learners trained over different historical horizons. Our results show that online learning reduces prediction error by 80% compared to a static model. The online ensemble and the proposed uncertainty-guided ensemble further reduce error by approximately 6%, and 10% respectively, relative to standard single-model online learning, while also providing calibrated uncertainty estimates to support operational decision-making.
AI↗
EFIT-Prime: Probabilistic and physics-constrained reduced-order neural network model for equilibrium reconstruction in DIII-D
We introduce EFIT-Prime, a novel machine learning surrogate model for EFIT (Equilibrium FIT) that integrates probabilistic and physics-informed methodologies to overcome typical limitations associated with deterministic and ad hoc neural network architectures. EFIT-Prime utilizes a neural architecture search-based deep ensemble for robust uncertainty quantification, providing scalable and efficient neural architectures that comprehensively quantify both data and model uncertainties. Physically informed by the Grad–Shafranov equation, EFIT-Prime applies a constraint on the current density J tor and a smoothness constraint on the first derivative of the poloidal flux, ensuring physically plausible solutions. Furthermore, the spatial location of the diagnostics is explicitly incorporated in the inputs to account for their spatial correlation. Extensive evaluations demonstrate EFIT-Prime's accuracy and robustness across diverse scenarios, most notably showing good generalization on negative-triangularity discharges that were excluded from training. Timing studies indicate an ensemble inference time of 15 ms for predicting a new equilibrium, offering the possibility of plasma control in real-time, if the model is optimized for speed.
70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
The CluMPR galaxy cluster-finding algorithm and DESI legacy survey galaxy cluster catalogue
ABSTRACT Galaxy clusters enable unique opportunities to study cosmology, dark matter, galaxy evolution, and strongly lensed transients. We here present a new cluster-finding algorithm, CluMPR (Clusters from Masses and Photometric Redshifts), that exploits photometric redshifts (photo-z’s) as well as photometric stellar mass measurements. CluMPR uses a 2D binary search tree to search for overdensities of massive galaxies with similar redshifts on the sky and then probabilistically assigns cluster membership by accounting for photo-z uncertainties. We leverage the deep DESI Legacy Survey grzW1W2 imaging over one-third of the sky to create a catalogue of $\sim 300\, 000$ galaxy cluster candidates out to z = 1, including tabulations of member galaxies and estimates of each cluster’s total stellar mass. Compared to other methods, CluMPR is particularly effective at identifying clusters at the high end of the redshift range considered (z = 0.75–1), with minimal contamination from low-mass groups. These characteristics make it ideal for identifying strongly lensed high-redshift supernovae and quasars that are powerful probes of cosmology, dark matter, and stellar astrophysics. As an example application of this cluster catalogue, we present a catalogue of candidate wide-angle strongly lensed quasars in Appendix C. The nine best candidates identified from this sample include two known lensed quasar systems and a possible changing-look lensed QSO with SDSS spectroscopy. All code and catalogues produced in this work are publicly available (see Data Availability).
79 ASTRONOMY AND ASTROPHYSICS↗
Field scale application of deep learning-based data assimilation for parameter estimation and uncertainty quantification at a geologic carbon storage site
Explore the source record for details and available documents.
Yoon, Hongkyu↗
The role of ProtoDUNE-SP in future oscillation physics
Long-baseline oscillation experiments are entering a new phase where neutrino beams can provide ultra-high fuxes over long distances. These beams will lead to an era of high statistics in large neutrino detectors, creating new systematic uncertainty-limited analyses. The Deep Underground Neutrino Experiment (DUNE) plans a high fux beam to travel 1300 km to 10-kT liquid argon detector modules. While neither the near detector nor the far detector has been constructed, prototypes of the DUNE Far Detector have been built and operated at CERN. One prototype, ProtoDUNE Single-Phase, is a full-scale mockup of the first 10-kT DUNE Far Detector module and can inform DUNE on detector performance and particle passage in argon. The proceedings discuss some published and in-progress results from ProtoDUNE-SP and how these results can impact current sensitivity studies and future analyses.
Diurba, Richard↗
Deep Space Network Scheduling Using Multi-Objective Optimization with Uncertainty
We have developed a novel technique to incorporate uncertainty modeling within an evolutionary algorithm approach to multi-objective scheduling, with the goal of identifying a Pareto frontier (tradeoff curve) that recognizes the likelihood of events that can impact the schedule outcome. Our approach is particularly applicable to the generation of multiobjective optimized robust schedules, where objectives are assigned a service level, for example that we require an objective value to be greater than or equal to X with Y% confidence. We have demonstrated that such an approach can, for example, minimize scheduling on less reliable resources, based solely on a resource reliability model and not on any ad hoc heuristics. We have also investigated an alternative method of optimizing for robustness, in which we add to the set of objectives a failure risk objective to minimize. We compare the advantages and disadvantages of these two approaches. Future plans for further developing this technology include its application to space-based observatory scheduling problems.
Johnston, Mark D.↗
Validation of Globsnow-2 Snow Water Equivalent Over Eastern Canada
In Qubec, Eastern Canada, snowmelt runoff contributes more than 30% of the annual energy reserve for hydroelectricity production, and uncertainties in annual maximum snow water equivalent (SWE) over the region are one of the main constraints for improved hydrological forecasting. Current satellite-based methods for mapping SWE over Qubec's main hydropower basins do not meet Hydro-Qubec operational requirements for SWE accuracies with less than 15% error. This paper assesses the accuracy of the GlobSnow-2 (GS-2) SWE product, which combines microwave satellite data and in situ measurements, for hydrological applications in Qubec. GS-2 SWE values for a 30-year period (1980 to 2009) were compared with space- and time-matched values from a comprehensive dataset of in situ SWE measurements (a total of 38,990 observations in Eastern Canada). The root mean square error (RMSE) of the GS-2 SWE product is 94.1+/- 20.3 mm, corresponding to an overall relative percentage error (RPE) of 35.9%. The main sources of uncertainty are wet and deep snow conditions (when SWE is higher than 150 mm), and forest cover type. However, compared to a typical stand-alone brightness temperature channel difference algorithm, the assimilation of surface information in the GS-2 algorithm clearly improves SWE accuracy by reducing the RPE by about 30%. Comparison of trends in annual mean and maximum SWE between surface observations and GS-2 over 1980-2009 showed agreement for increasing trends over southern Qubec, but less agreement on the sign and magnitude of trends over northern Qubec. Extended at a continental scale, the GS-2 SWE trends highlight a strong regional variability.
Snow water equivalent↗
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↗
Two Problems in Applied Stochastic Processes: Scaling Limits of the Multiple Range in 1D Random Walk and Uncertainty Quantification in Satellite Imagery
Explore the source record for details and available documents.
97 MATHEMATICS AND COMPUTING↗
Uncertainty Quantification in Satellite Imagery
This is a presentation for Conner's Dissertation Defense which will be help on 2024/12/12, based on the work he has done with the NNSS.
97 MATHEMATICS AND COMPUTING↗
Ensemble learning-iterative training machine learning for uncertainty quantification and automated experiment in atom-resolved microscopy
Deep learning has emerged as a technique of choice for rapid feature extraction across imaging disciplines, allowing rapid conversion of the data streams to spatial or spatiotemporal arrays of features of interest. However, applications of deep learning in experimental domains are often limited by the out-of-distribution drift between the experiments, where the network trained for one set of imaging conditions becomes sub-optimal for different ones. This limitation is particularly stringent in the quest to have an automated experiment setting, where retraining or transfer learning becomes impractical due to the need for human intervention and associated latencies. Here we explore the reproducibility of deep learning for feature extraction in atom-resolved electron microscopy and introduce workflows based on ensemble learning and iterative training to greatly improve feature detection. This approach allows incorporating uncertainty quantification into the deep learning analysis and also enables rapid automated experimental workflows where retraining of the network to compensate for out-of-distribution drift due to subtle change in imaging conditions is substituted for human operator or programmatic selection of networks from the ensemble. This methodology can be further applied to machine learning workflows in other imaging areas including optical and chemical imaging.
36 MATERIALS SCIENCE↗