Error-in-variables modelling for operator learning benchmarks
Explore the source record for details and available documents.
Patel, Ravi Ghanshyam↗
SEARCH · Engineering Papers
Results for “error modeling”
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.
Q-BEEP: Quantum Bayesian Error Mitigation Employing Poisson Modeling over the Hamming Spectrum
Quantum computing technology has grown rapidly in recent years, with new technologies being explored, error rates being reduced, and quantum processor’s qubit capacity growing. However, near-term quantum algorithms are still unable to be induced without compounding consequential levels of noise, leading to non-trivial erroneous results. Quantum Error Correction (in-situ error mitigation) and Quantum Error Mitigation (post-induction error mitigation) are promising fields of research within the quantum algorithm scene, aiming to alleviate quantum errors, increasing the overall fidelity and hence the overall quality of circuit induction. Earlier this year, a pioneering work, namely HAMMER, published in ASPLOS-22 demonstrated the existence of a latent structure regarding post-circuit induction errors when mapping to the Hamming spectrum. However, they intuitively assumed that errors occur in local clusters, and that at higher average Hamming distances this structure falls away. In this work, we show that such a correlation structure is not only local but extends certain non-local clustering patterns which can be precisely described by a Poisson distribution model taking the input circuit, the device run time status (i.e., calibration statistics) and qubit topology into consideration. Using this quantum error characterizing model, we developed an iterative algorithm over the generated Bayesian network state-graph for post-induction error mitigation. Thanks to more precise modeling of the error distribution latent structure and the new iterative method, our Q-Beep approach provides state of the art performance and can boost circuit execution fidelity by up to 234.6% on Bernstein-Vazirani circuits and on average 71.0% on QAOA solution quality, using 16 practical IBMQ quantum processors. For other benchmarks such as those in QASMBench, the fidelity improvement is up to 17.8%. Q-Beep is a light-weight post-processing technique that can be performed offline and remotely, making it a useful tool for quantum vendors to integrate and provide more reliable circuit induction results.
Stein, Samuel A.↗
Error analysis of low-fidelity models for wake steering based on field measurements
The observations collected by two scanning lidars deployed on the roof of a 2.8-MW turbine undergoing a series of imposed yaw offsets are analyzed. The wake lateral displacement detected by the rear-facing lidar correlates well with the yaw offset sensed by the forward-facing lidar. We find that the high-frequency part of the yaw offset signal is connected to wake meandering, whereas the low frequency component is a good predictor for wake displacement due to yaw misalignment. Conditionally averaged wake velocity data for different yaw offsets are used as benchmarks for the validation of a linearized Reynolds-averaged Navier-Stokes and an empirical wake model. A mean error as low as 2% and a good prediction of the wake trajectory are achieved, provided that the wake recovery rate matches the observations.
17 WIND ENERGY↗
Back to Basics: First Principles of HRA
Human Reliability Analysis (HRA) continues to suffer from a myriad of technical gaps in its foundations. Core issues have been widely discussed but remain largely unaddressed, although the path taken to this point was paved with good intentions. Now, while this characterization may seem dramatic it is not intended to throw doubt onto HRA as a discipline or to diminish the value it has on increasing safety and resilience in nuclear power. Rather, it is intended to communicate that HRA has some serious challenges and is desperately in need of a critical evaluation. To quote Dougherty [1], “the misrepresentations that will undoubtedly arise may serve to provoke the necessary creativity.” What bears noting is that in the last thirty years since Alan Swain declared that “no adequate HRA had ever been done,” there has been little effort to correct the issues he highlighted. Human factors practice is still poorly implemented, many complex systems are fundamentally incompatible with reliable human performance, and there remains poor foundational support for many of the assumptions within HRA. We propose three categories of issues in HRA, philosophical, theoretical, and empirical and potential solutions to these issues.
22 GENERAL STUDIES OF NUCLEAR REACTORS↗
Multi-fidelity thermal modeling of laser powder bed additive manufacturing
Laser powder bed fusion (LPBF) Additive manufacturing (AM) has attracted interest as an agile method of building production metal parts to reduce design-build-test cycle times for systems. However, predicting part performance is difficult due to inherent process variabilities. This makes qualification challenging. Computational process models have attempted to address some of these challenges, including mesoscale, full physics models and reduced fidelity conduction models. The goal of this work is credible multi-fidelity modeling of the LPBF process by investigating methods for estimating the error between models of two different fidelities. Two methods of error estimation are investigated, adjoint-based error estimation and Bayesian calibration. Adjoint-based error estimation is found to effectively bounding the error between the two models, but with very conservative bounds, making predictions highly uncertain. Bayesian parameter calibration applied to conduction model heat source parameters is found to effectively bound the observed error between the models for melt pool morphology quantities of interest. However, the calibrations do not effectively bound the error in heat distribution.
36 MATERIALS SCIENCE↗
Machine learning materials properties with accurate predictions, uncertainty estimates, domain guidance, and persistent online accessibility
One compelling vision of the future of materials discovery and design involves the use of machine learning (ML) models to predict materials properties and then rapidly find materials tailored for specific applications. However, realizing this vision requires both providing detailed uncertainty quantification (model prediction errors and domain of applicability) and making models readily usable. At present, it is common practice in the community to assess ML model performance only in terms of prediction accuracy (e.g. mean absolute error), while neglecting detailed uncertainty quantification and robust model accessibility and usability. Here, we demonstrate a practical method for realizing both uncertainty and accessibility features with a large set of models. We develop random forest ML models for 33 materials properties spanning an array of data sources (computational and experimental) and property types (electrical, mechanical, thermodynamic, etc). All models have calibrated ensemble error bars to quantify prediction uncertainty and domain of applicability guidance enabled by kernel-density-estimate-based feature distance measures. All data and models are publicly hosted on the Garden-AI infrastructure, which provides an easy-to-use, persistent interface for model dissemination that permits models to be invoked with only a few lines of Python code. We demonstrate the power of this approach by using our models to conduct a fully ML-based materials discovery exercise to search for new stable, highly active perovskite oxide catalyst materials.
domain of applicability↗
Active learning using hybrid surrogate tool life modeling for machining process optimization
Here, this paper describes an active learning approach for part-to-part iterative machining process optimization using a hybrid surrogate tool life model. A probabilistic interpolating tool life model is developed by combining the empirical Taylor-type tool life equation and the model fit error. The probabilistic tool life model is then used to calculate the machining cost per part distribution. The optimal machining parameters are selected using an expected improvement in machining cost per part criterion. The method is validated numerically using experimental results; the results show a median convergence error of 2.2% after three tests over 400 simulations. The method is validated experimentally on two industrial applications for Ti-6Al-4V roughing resulting in a cost per part reduction greater than 23% after two tests. The described method is a robust solution for rapid convergence to optimal machining parameters in an industrial production environment.
Active learning↗
Effects of natural zeolites on field-scale geologic noble gas transport
Improving predictive models for noble gas transport through natural materials at the field-scale is an essential component of improving US nuclear monitoring capabilities. Several field-scale experiments with a gas transport component have been conducted at the Nevada National Security Site (Non-Proliferation Experiment, Underground Nuclear Explosion Signatures Experiment). However, the models associated with these experiments have not treated zeolite minerals as gas adsorbing phases. This is significant as zeolites are a common alteration mineral with a high abundance at these field sites and are shown here to significantly fractionate noble gases during field-scale transport. This fractionation and associated retardation can complicate gas transport predictions by reducing the signal-to-noise ratio to the detector (e.g. mass spectrometers or radiation detectors) enough to mask the signal or make the data difficult to interpret. Omitting adsorption-related retardation data of noble gases in predictive gas transport models therefore results in systematic errors in model predictions where zeolites are present.Herein is presented noble gas adsorption data collected on zeolitized and non-zeolitized tuff. Experimental results were obtained using a unique piezometric adsorption system designed and built for this study. Data collected were then related to pure-phase mineral analyses conducted on clinoptilolite, mordenite, and quartz. These results quantify the adsorption capacity of materials present in field-scale systems, enabling the modeling of low-permeability rocks as significant sorption reservoirs vital to bulk transport predictions.
58 GEOSCIENCES↗
Model Evaluation by Measurements from Collocated Remote Sensors in Complex Terrain
Model improvement efforts involve an evaluation of changes in model skill in response to changes in model physics and parameterization. When using wind measurements from various remote sensors to determine model forecast accuracy, it is important to understand the effects of measurement-uncertainty differences among the sensors resulting from differences in the methods of measurement, the vertical and temporal resolution of the measurements, and the spatial variability of these differences. Here we quantify instrument measurement variability in 80-m wind speed during WFIP2 and its impact on the calculated errors and the change in error from one model version to another. Here the model versions tested involved updates in model physics from HRRRv1 to HRRRv4, and reductions in grid interval from 3 km to 750 m. Model errors were found to be 2–3 m s –1 . Differences in errors as determined by various instruments at each site amounted to about 10% of this value, or 0.2–0.3 m s –1 . Changes in model skill due to physics or grid-resolution updates also differed depending on the instrument used to determine the errors; most of the instrument-to-instrument differences were ~0.1 m s –1 , but some reached 0.3 m s –1 . All instruments at a given site mostly showed consistency in the sign of the change in error. In two examples, though, the sign changed, illustrating a consequence of differences in measurements: errors determined using one instrument may show improvement in model skill, whereas errors determined using another instrument may indicate degradation. This possibility underscores the importance of having accurate measurements to determine the model error.
54 ENVIRONMENTAL SCIENCES↗
Evaluating skill in predicting the Interdecadal Pacific Oscillation in initialized decadal climate prediction hindcasts in E3SMv1 and CESM1 using two different initialization methods and a small set of start years
Abstract It is a daunting challenge to conduct initialized hindcasts with enough ensemble members and associated start years to form a drifted climatology from which to compute the anomalies necessary to quantify the skill of the hindcasts when compared to observations. This limits the ability to experiment with case studies and other applications where only a few initial years are needed. Here we run a set of hindcasts with CESM1 and E3SMv1 using two different initialization methods for a limited set of start years and use the respective uninitialized free-running historical simulations to form the model climatologies. Since the drifts from the observed initial states in the hindcasts toward the uninitialized model state are large and rapid, after a few years the drifted initialized models approach the uninitialized model climatological errors. Therefore, hindcasts from the limited start years can use the uninitialized climatology to represent the drifted model states after about lead year 3, providing a means to compute forecast anomalies in the absence of a large hindcast sample. There is comparable skill for predicting spatial patterns of multi-year Pacific sea surface temperature anomalies in the domain of the Interdecadal Pacific Oscillation using this method compared to the conventional methodology with a large hindcast data set, though there is a model dependence to the drifts in the two initialization methods.
54 ENVIRONMENTAL SCIENCES↗
Multiscale analysis in solids with unseparated scales: fine-scale recovery, error estimation, and coarse-scale adaptivity
There are several engineering applications in which the assumptions of homogenization and scale separation may be violated, in particular, for metallic structures constructed through additive manufacturing. Instead of resorting to direct numerical simulation of the macroscale system with an embedded fine scale, an alternative approach is to use an approximate macroscale constitutive model, but then estimate the model-form error using a posteriori error estimation techniques and subsequently adapt the macroscale model to reduce the error for a given boundary value problem and quantity of interest. Here, we investigate this approach to multiscale analysis in solids with unseparated scales using the example of an additively manufactured metallic structure consisting of a polycrystalline microstructure that is neither periodic nor statistically homogeneous. As a first step to the general nonlinear case, we focus here on linear elasticity in which each grain within the polycrystal is linear elastic but anisotropic.
42 ENGINEERING↗
Characterizing Impacts of Storage Faults on HPC Applications: A methodology and insights
In recent years, the increasing complexity in scientific simulations and emerging demands for training heavy artificial intelligence models require massive and fast data accesses, which urges high-performance computing (HPC) platforms to equip with more advanced storage infrastructures such as solid-state disks (SSDs). While SSDs offer high-performance I/O, it remains unclear about the reliability challenges faced by the HPC applications under the SSD-related failures, in particular, failures resulting in data corruptions. The goal of this paper is to understand the impact of SSD-related data corruptions on the behaviors of complex HPC applications. To this end, we propose FFIS, a FUSE-based fault injection framework that systematically introduces storage faults into the application layer to model the errors originated from SSDs. FFIS is able to plant different I/O related faults into the data returned from underlying file systems, which also enables the investigation on the error resilience characteristics of the scientific file format for the first time. We demonstrate the use of FFIS with three representative real HPC applications, show how each application reacts to the data corruptions, and provide insights on the error resilience of the widely-adopted HDF5 file format for the HPC applications.
Fang, Bo↗
Modeling noise in global Mølmer-Sørensen interactions applied to quantum approximate optimization
Many-qubit Mølmer-Sørensen (MS) interactions applied to trapped ions offer unique capabilities for quantum information processing, with applications including quantum simulation and the quantum approximate optimization algorithm (QAOA). Here, we develop a physical model to describe many-qubit MS interactions under four sources of experimental noise: vibrational mode frequency fluctuations, laser power fluctuations, thermal initial vibrational states, and state preparation and measurement errors. The model parametrizes these errors from simple experimental measurements, without free parameters. Herein, we validate the model in comparison with experiments that implement sequences of MS interactions on two 171 Yb + ions. The model shows reasonable agreement after several MS interactions as quantified by the reduced chi-squared statistic $χ^{2}_{red}$≈2. As an application we examine MaxCut QAOA experiments on three and six ions. The experimental performance is quantified by approximation ratios that are 91% and 83% of the optimal theoretical values. Our model predicts $0.93^{+0.03}_{-0.02}$ and $0.95^{+0.04}_{-0.03}$, respectively, with disagreement in the latter value attributable to secondary noise sources beyond those considered in our analysis. With realistic experimental improvements to reduce measurement error and radial trap frequency variations, the model achieves approximation ratios that are 99% of the optimal. Incorporating these improvements into future experiments is expected to reveal new aspects of noise for future modeling and experimental improvements.
71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
Multiphenomenology explosion monitoring (MultiPEM): a general framework for data interpretation and yield estimation
SUMMARY An underground nuclear explosion (UNE) couples mechanical energy into crustal rock, which propagates as seismic and acoustic waves. These different physical phenomena transport, by different pathways, to standoff detectors at varying distances. The transport pathways attenuate the original signal but in different ways. Enabled by correct statistical weighting, signal attenuation models can be used to combine these disparate sensor data to estimate the yield of an UNE. Contemporaneous statistical models, used in yield estimation, can be improved with an advanced partition of error for these physical signal propagation models. We present an advanced multivariate approach to error modelling of multiphenomenology physical signatures. In addition to measurement error, our error model represents physical model biases as random with a physics-based covariance structure. To illustrate this proposed framework, we demonstrate the estimation of explosion yield using openly available seismic and acoustic data from chemical single-point explosions.
Williams, Brian J.↗
Assessing Cognitive Impacts of Errors from Machine Learning and Deep Learning Models: Final Report
Due to their recent increases in performance, machine learning and deep learning models are being increasingly adopted across many domains for visual processing tasks. One such domain is international nuclear safeguards, which seeks to verify the peaceful use of commercial nuclear energy across the globe. Despite recent impressive performance results from machine learning and deep learning algorithms, there is always at least some small level of error. Given the significant consequences of international nuclear safeguards conclusions, we sought to characterize how incorrect responses from a machine or deep learning-assisted visual search task would cognitively impact users. We found that not only do some types of model errors have larger negative impacts on human performance than other errors, the scale of those impacts change depending on the accuracy of the model with which they are presented and they persist in scenarios of evenly distributed errors and single-error presentations. Further, we found that experiments conducted using a common visual search dataset from the psychology community has similar implications to a safeguards- relevant dataset of images containing hyperboloid cooling towers when the cooling tower images are presented to expert participants. While novice performance was considerably different (and worse) on the cooling tower task, we saw increased novice reliance on the most challenging cooling tower images compared to experts. These findings are relevant not just to the cognitive science community, but also for developers of machine and deep learning that will be implemented in multiple domains. For safeguards, this research provides key insights into how machine and deep learning projects should be implemented considering their special requirements that information not be missed.
97 MATHEMATICS AND COMPUTING↗