Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Learning with errors”

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 307 records · Page 17
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Detecting Anomalies in Time Series Using Kernel Density Approaches

This paper introduces a novel anomaly detection approach tailored for time series data with exclusive reliance on normal events during training. Our key innovation lies in the application of kernel-density estimation (KDE) to scrutinize reconstruction errors, providing an empirically derived probability distribution for normal events post-reconstruction. This non-parametric density estimation technique offers a nuanced understanding of anomaly detection, differentiating it from prevalent threshold-based mechanisms in existing methodologies. In post-training, events are encoded, decoded, and evaluated against the estimated density, providing a comprehensive notion of normality. In addition, we propose a data augmentation strategy involving variational autoencoder-generated events and a smoothing step for enhanced model robustness. The significance of our autoencoder-based approach is evident in its capacity to learn normal representation without prior anomaly knowledge. Through the KDE step on reconstruction errors, our method addresses the versatility of anomalies, departing from assumptions tied to larger reconstruction errors for anomalous events. Our proposed likelihood measure then distinguishes normal from anomalous events, providing a concise yet comprehensive anomaly detection solution. The extensive experimental results support the feasibility of our proposed method, yielding significantly improved classification performance by nearly 10% on the UCR benchmark data.

Frehner, Robin↗
DOE OSTI2021

Using Verified Lifting to Optimize Legacy Stencil Codes (Final Project Report)

This project investigated new techniques for compiling stencil and stencil-like computations. Stencils computations are commonly found in applications such as image processing, physical simulations, image processing, and machine learning. In recent years, many high-performance domain-specific languages (DSLs) have been proposed to optimize stencil computations. To leverage such DSLs, however, existing codes often need to be rewritten. Such rewriting is manual, labor intensive, and error-prone. To alleviate such issues, this project investigated the application of program synthesis and artificial learning techniques to enable stencil computations to automatically leverage new high-performance DSLs. Rather than constructing syntax driven rules, verified lifting uses program synthesis to search for a target code fragment to compile the given input code into. In addition, it also searches for a proof that validates how the found target code fragment preserves the semantics of the original input. Thus, the target code fragment is guaranteed to be semantically equivalent to the input.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

LSTM-Based Data Integration to Improve Snow Water Equivalent Prediction and Diagnose Error Sources

Accurate prediction of snow water equivalent (SWE) can be valuable for water resource managers. Recently, deep learning methods such as long short-term memory (LSTM) have exhibited high accuracy in simulating hydrologic variables and can integrate lagged observations to improve prediction, but their benefits were not clear for SWE simulations. Here we tested an LSTM network with data integration (DI) for SWE in the western United States to integrate 30-day-lagged or 7-day-lagged observations of either SWE or satellite-observed snow cover fraction (SCF) to improve future predictions. SCF proved beneficial only for shallow-snow sites during snowmelt, while lagged SWE integration significantly improved prediction accuracy for both shallow- and deep-snow sites. The median Nash–Sutcliffe model efficiency coefficient (NSE) in temporal testing improved from 0.92 to 0.97 with 30-day-lagged SWE integration, and root-mean-square error (RMSE) and the difference between estimated and observed peak SWE values d max were reduced by 41% and 57%, respectively. DI effectively mitigated accumulated model and forcing errors that would otherwise be persistent. Moreover, by applying DI to different observations (30-day-lagged, 7-day-lagged), we revealed the spatial distribution of errors with different persistent lengths. For example, integrating 30-day-lagged SWE was ineffective for ephemeral snow sites in the southwestern United States, but significantly reduced monthly-scale biases for regions with stable seasonal snowpack such as high-elevation sites in California. These biases are likely attributable to large interannual variability in snowfall or site-specific snow redistribution patterns that can accumulate to impactful levels over time for nonephemeral sites. These results set up benchmark levels and provide guidance for future model improvement strategies.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Learning to Predict Arbitrary Quantum Processes

We present an efficient machine-learning (ML) algorithm for predicting any unknown quantum process ℰ over 𝑛 qubits. For a wide range of distributions 𝒟 on arbitrary 𝑛-qubit states, we show that this ML algorithm can learn to predict any local property of the output from the unknown process ℰ, with a small average error over input states drawn from 𝒟. The ML algorithm is computationally efficient even when the unknown process is a quantum circuit with exponentially many gates. Our algorithm combines efficient procedures for learning properties of an unknown state and for learning a low-degree approximation to an unknown observable. The analysis hinges on proving new norm inequalities, including a quantum analogue of the classical Bohnenblust-Hille inequality, which we derive by giving an improved algorithm for optimizing local Hamiltonians. Numerical experiments on predicting quantum dynamics with evolution time up to 10 6 and system size up to 50 qubits corroborate our proof. Overall, our results highlight the potential for ML models to predict the output of complex quantum dynamics much faster than the time needed to run the process itself.

quantum computation↗
DOE OSTI2021

Validation of Jezebel Reactivity Coefficients and Sensitivity Analysis

Nuclear data validation is often performed today using criticality measurements. The gold standard for criticality measurements is the International Criticality Safety Benchmark Experiment Project (ICSBEP). The validation specifically focuses on the effective multiplication factor ($k_{eff}$). $K_{eff}$ is a relatively easy parameter to infer and has reduced uncertainty due to being at or above critical. However, while $k_{eff}$ is the most documented parameter and its uncertainties and sensitivities have been evaluated in great detail, it cannot be used as a standalone metric to determine inaccuracies in nuclear data (e.g., cross section data, PFNS, nu), which is based on theory, physics, and differential measurements. The Experiments Underpinned by Computational Learning for Improvements in nuclear Data (EUCLID) project aims to identify compensating errors in specific isotope nuclear data by optimally designing experiments that are, or are not sensitive to a suite of measurement parameters beyond $k_{eff}$. By identifying parameters that are sensitive to each other, oppositely sensitive, or have substantial magnitude differences in sensitivity, experiments can be designed to constrain questionable nuclear data. One sensitivity that is of particular interest to this project includes the sensitivity of reactivity coefficients. Reactivity coefficients compare reactivity, which is related to $k_{eff}$ at two different states therefore being sensitive to small changes in the system. The most common type of reactivity coefficient measurements is comparison to void for a small sample within the assembly. It is key that the sample sizes are small enough to not affect the flux of the full system. Reactivity coefficients were evaluated for many early experiments to better understand transport corrected cross sections. In fact, ICSBEP includes reactivity coefficient results as “Supplemental Measurements” in appendices for a handful of older benchmarks. One of those benchmarks is Jezebel, the bare Pu critical assembly. This paper compares new simulations of reactivity coefficients for Jezebel, and explores the sensitivity of reactivity coefficients to small changes in nuclear data.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2023

A deep learning model for automatic analysis of cavities in irradiated materials

Transmission electron microscopy (TEM) is a commonly used technique in materials science for defect investigation. Quantitative analysis of defects is important for understanding the properties of a material, but manual analysis of TEM micrographs can be time-consuming and prone to error, especially when the defects have irregular shapes rather than spherical shapes. Many existing methods or deep learning models do not handle a wide range of sizes for the same object type within a single image. In this work, we present a framework that enables users to train an instance segmentation model called Mask R- CNN on any microstructure dataset, perform multi-detection on the same image at different scales, and obtain properties (e.g., size, area) of the objects based on the desired shape (e.g., circle, ellipse, rectangle). Additionally, we have developed a parallel detection module that uses multiple GPUs to increase the efficiency of the object detection process. We demonstrate the capabilities of our framework using a set of TEM images of cavities with different shapes, size distributions, and background contrasts. Finally, we show that the performance of our model in terms of density, size, and swelling of the cavities is comparable to the human average and that our model achieves the highest recall value compared to existing methods due to the use of image multi-rescaling.

36 MATERIALS SCIENCE↗
DOE OSTI2022

PRODeepSyn: predicting anticancer synergistic drug combinations by embedding cell lines with protein–protein interaction network

Abstract Although drug combinations in cancer treatment appear to be a promising therapeutic strategy with respect to monotherapy, it is arduous to discover new synergistic drug combinations due to the combinatorial explosion. Deep learning technology holds immense promise for better prediction of in vitro synergistic drug combinations for certain cell lines. In methods applying such technology, omics data are widely adopted to construct cell line features. However, biological network data are rarely considered yet, which is worthy of in-depth study. In this study, we propose a novel deep learning method, termed PRODeepSyn, for predicting anticancer synergistic drug combinations. By leveraging the Graph Convolutional Network, PRODeepSyn integrates the protein–protein interaction (PPI) network with omics data to construct low-dimensional dense embeddings for cell lines. PRODeepSyn then builds a deep neural network with the Batch Normalization mechanism to predict synergy scores using the cell line embeddings and drug features. PRODeepSyn achieves the lowest root mean square error of 15.08 and the highest Pearson correlation coefficient of 0.75, outperforming two deep learning methods and four machine learning methods. On the classification task, PRODeepSyn achieves an area under the receiver operator characteristics curve of 0.90, an area under the precision–recall curve of 0.63 and a Cohen’s Kappa of 0.53. In the ablation study, we find that using the multi-omics data and the integrated PPI network’s information both can improve the prediction results. Additionally, the case study demonstrates the consistency between PRODeepSyn and previous studies.

Wang, Xiaowen↗
DOE OSTI2023

Low-energy Electron-track Imaging for a Liquid Argon Time-projection-chamber Telescope Concept Using Probabilistic Deep Learning

The GammaTPC is an MeV-scale single-phase liquid argon time-projection-chamber gamma-ray telescope concept with a novel dual-scale pixel-based charge-readout system. It promises to enable a significant improvement in sensitivity to MeV-scale gamma rays over previous telescopes. The novel pixel-based charge readout allows for imaging of the tracks of electrons scattered by Compton interactions of incident gamma rays. The two primary contributors to the accuracy of a Compton telescope in reconstructing an incident gamma-ray’s original direction are its energy and position resolution. In this work, we focus on using deep learning to optimize the reconstruction of the initial position and direction of electrons scattered in Compton interactions, including using probabilistic models to estimate predictive uncertainty. We show that the deep-learning models are able to predict locations of Compton scatters of MeV-scale gamma rays from simulated 500 μm pixel-based data to better than 1 mm rms error and are sensitive to the initial direction of the scattered electron. We compare and contrast different deep-learning uncertainty estimation algorithms for reconstruction applications. Additionally, we show that event-by-event estimates of the uncertainty of the locations of the Compton scatters can be used to select those events that were reconstructed most accurately, leading to improvement in locating the origin of gamma-ray sources on the sky.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2022

Performance and power modeling and prediction using MuMMI and 10 machine learning methods

Energy-efficient scientific applications require insight into how high performance computing system features impact the applications' power and performance. This insight can result from the development of performance and power models. Here, in this article, we use the modeling and prediction tool MuMMI (Multiple Metrics Modeling Infrastructure) and 10 machine learning methods to model and predict performance and power consumption and compare their prediction error rates. We use an algorithm-based fault-tolerant linear algebra code and a multilevel checkpointing fault-tolerant heat distribution code to conduct our modeling and prediction study on the Cray XC40 Theta and IBM BG/Q Mira at Argonne National Laboratory and the Intel Haswell cluster Shepard at Sandia National Laboratories. Our experimental results show that the prediction error rates in performance and power using MuMMI are less than 10% for most cases. By utilizing the models for runtime, node power, CPU power, and memory power, we identify the most significant performance counters for potential application optimizations, and we predict theoretical outcomes of the optimizations. Based on two collected datasets, we analyze and compare the prediction accuracy in performance and power consumption using MuMMI and 10 machine learning methods.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Learning Many-Body Hamiltonians with Heisenberg-Limited Scaling

Learning a many-body Hamiltonian from its dynamics is a fundamental problem in physics. Here, in this Letter, we propose the first algorithm to achieve the Heisenberg limit for learning an interacting N-qubit local Hamiltonian. After a total evolution time of $\mathscr{O}$⁡(ε –1 ), the proposed algorithm can efficiently estimate any parameter in the N-qubit Hamiltonian to ε error with high probability. Our algorithm uses ideas from quantum simulation to decouple the unknown N-qubit Hamiltonian H into noninteracting patches and learns H using a quantum-enhanced divide-and-conquer approach. The proposed algorithm is robust against state preparation and measurement error, does not require eigenstates or thermal states, and only uses polylog⁡(ε –1 ) experiments. In contrast, the best existing algorithms require $\mathscr{O}$⁡⁡(ε –2 ) experiments and total evolution time. We prove a matching lower bound to establish the asymptotic optimality of our algorithm.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2022

Residential Demand Flexibility: Modeling Occupant Behavior using Sociodemographic Predictors

Demand flexibility (DF) has the potential to increase the saturation of renewables in the grid and reduce operating costs for both utilities and customers. However, less than 8% of U.S. residential electric customers are enrolled in DF programs. A major research gap on this topic is an uneven understanding of behavioral drivers of electricity use and DF program participation at the household level. In this study, we employ machine learning models to predict residential occupant behavior in activities relevant to DF. We model occupants' extensive decisions (i.e., choice of action) and intensive behaviors (i.e., amount of time spent) during peak and off-peak time periods using the publicly available American Time Use Survey, which includes activities data for approximately 200,000 respondents. In our machine learning models, predictions for both extensive and intensive behavior fell within a +/-20% error margin at the aggregate level. We identify 13 key sociodemographic predictors of DF-related intensive behavior using LASSO inference and beta coefficient ranking. However, these top predictors differ by activity, suggesting potential scope for differential user targeting for DF events and technologies during program design. This work also contributes to understanding when and who might adopt these DF technologies based on their daily routine activities.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2021

A novel implicit hybrid machine learning model and its application for reinforcement learning

A novel methodology to develop implicit hybrid models is presented. PyTorch is used to integrate physics-based equations with machine learning models. Automatic differentiation of the hybrid model is leveraged to solve the implicit equations. Iterative solving enables gradient based updates to the machine learning model. The novel methodology is compared to an explicit hybrid approach on a continuously stirred tank reactor (CSTR). The novel method results in a lower modelling error. Both hybrid models effectively train with noisy data. To test the implicit hybrid model, it is employed as a reinforcement learning (RL) training model. The RL algorithm trained on the hybrid model outperforms real time optimization of the CSTR and performs nearly as well as RL trained directly on the CSTR and a traditional gradient based approach. Training RL directly on the CSTR requires over 60,000 system interactions compared to 6000 historical data points for hybrid model development.

42 ENGINEERING↗
DOE OSTI2024

Understanding Mixed Precision GEMM with MPGemmFI: Insights into Fault Resilience

Emerging deep learning workloads urgently need fast general matrix multiplication (GEMM). Thus, one of the critical features of machine-learning-specific accelerators such as NVIDIA Tensor Cores, AMD Matrix Cores, and Google TPUs is the support of mixed-precision enabled GEMM. For DNN models, lower-precision FP data formats and computation offer acceptable correctness but significant performance, area, and memory footprint improvement. While promising, the mixed-precision computation on error resilience remains unexplored. To this end, we develop a fault injection framework that systematically injects fault into the mixed-precision computation results. We investigate how the faults affect the accuracy of machine learning applications. Based on the characteristics of error resilience, we offer lightweight error detection and correction solutions that significantly improve the overall model accuracy by 75% if the models experience hardware faults. The solutions can be efficiently integrated into the accelerator's pipelines.

Fang, Bo↗
DOE OSTI2023

Physics‐Informed Machine Learning for Inverse Design of Optical Metamaterials

Optical metamaterials manipulate light through various confinement and scattering processes, offering unique advantages like high performance, small form factor and easy integration with semiconductor devices. However, designing metasurfaces with suitable optical responses for complex metamaterial systems remains challenging due to the exponentially growing computation cost and the ill‐posed nature of inverse problems. To expedite the computation for the inverse design of metasurfaces, a physics‐informed deep learning (DL) framework is used. A tandem DL architecture with physics‐based learning is used to select designs that are scientifically consistent, have low error in design prediction, and accurate reconstruction of optical responses. The authors focus on the inverse design of a representative plasmonic device and consider the prediction of design for the optical response of a single wavelength incident or a spectrum of wavelength in the visible light range. The physics‐based constraint is derived from solving the electromagnetic wave equations for a simplified homogenized model. The model converges with an accuracy up to 97% for inverse design prediction with the optical response for the visible light spectrum as input, and up to 96% for optical response of single wavelength of light as input, with optical response reconstruction accuracy of 99%.

Sarkar, Sulagna↗
DOE OSTI2025

Adaptive Algebraic Derivative Estimation for Battery Electric Buses Energy Consumption Forecasting

The limited service life of onboard batteries for EVs is a challenge, underscoring the need for real-time battery usage prediction. This paper proposes an adaptive Algebraic Derivative Estimation (ADE) approach for forecasting the energy consumption of battery electric buses. By dynamically adjusting the sliding window length, the adaptive ADE retains the fixed-length ADE’s key advantage—namely, operating online without reliance on extensive historical datasets—while substantially bolstering forecast accuracy by actively trading estimation bias off estimation variance. Comparative experiments against both the conventional ADE with a fixed length and a representative machine learning algorithm, XGBoost, were conducted, with performance evaluated via root mean square error, mean absolute error, and the coefficient of determination. The results demonstrate that the proposed approach significantly outperforms baseline methods.

Cui, Tianyang [The University of Texas at Dallas]↗
DOE OSTI2025

Periodic GFN1-xTB Tight Binding: A Generalized Ewald Partitioning Scheme for the Klopman–Ohno Function

A novel formulation is presented for the treatment of electrostatics in the periodic GFN1-xTB tight-binding model. Periodic GFN1-xTB is hindered by the functional form of the second-order electrostatics, which only recovers Coulombic behavior at large interatomic distances and lacks a closed-form solution for its Fourier transform. We address this by introducing a binomial expansion of the Klopman–Ohno function to partition short- and long-range interactions, enabling the use of a generalized Ewald summation for the solution of the electrostatic energy. This approach is general and is applicable to any damped potential of the form |R n + c| –m . Benchmarks on the X23 molecular crystal dataset and a range of prototypical bulk semiconductors demonstrate that this systematic treatment of the electrostatics eliminates unphysical behavior in the equation of state curves. In the bulk systems studied, we observe a mean absolute error in total energy of 35 meV/atom, comparable to the machine-learned universal force field, M3GNet, and sufficiently precise for structure relaxation. These results highlight the promising potential of GFN1-xTB as a universal tight-binding parametrization.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Machine learning for impurity charge-state transition levels in semiconductors from elemental properties using multi-fidelity datasets

Quantifying charge-state transition energy levels of impurities in semiconductors is critical to understanding and engineering their optoelectronic properties for applications ranging from solar photovoltaics to infrared lasers. While these transition levels can be measured and calculated accurately, such efforts are time-consuming and more rapid prediction methods would be beneficial. Here, we significantly reduce the time typically required to predict impurity transition levels using multi-fidelity datasets and a machine learning approach employing features based on elemental properties and impurity positions. We use transition levels obtained from low-fidelity (i.e., local-density approximation or generalized gradient approximation) density functional theory (DFT) calculations, corrected using a recently proposed modified band alignment scheme, which well-approximates transition levels from high-fidelity DFT (i.e., hybrid HSE06). Further, the model fit to the large multi-fidelity database shows improved accuracy compared to the models trained on the more limited high-fidelity values. Crucially, in our approach, when using the multi-fidelity data, high-fidelity values are not required for model training, significantly reducing the computational cost required for training the model. Our machine learning model of transition levels has a root mean squared (mean absolute) error of 0.36 (0.27) eV vs high-fidelity hybrid functional values when averaged over 14 semiconductor systems from the II–VI and III–V families. As a guide for use on other systems, we assessed the model on simulated data to show the expected accuracy level as a function of bandgap for new materials of interest. Finally, we use the model to predict a complete space of impurity charge-state transition levels in all zinc blende III–V and II–VI systems.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗