Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Inference”

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

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

At least 1,729 records · Page 96

Scalable in-memory RDFS closure on billions of triples.

We present an RDFS closure algorithm, specifically designed and implemented on the Cray XMT supercomputer, that obtains inference rates of 13 million inferences per second on the largest system configuration we used. The Cray XMT, with its large global memory (4TB for our experiments), permits the construction of a conceptually straightforward algorithm, fundamentally a series of operations on a shared hash table. Each thread is given a partition of triple data to process, a dedicated copy of the ontology to apply to the data, and a reference to the hash table into which it inserts inferred triples. The global nature of the hash table allows the algorithm to avoid a common obstacle for distributed memory machines: the creation of duplicate triples. On LUBM data sets ranging between 1.3 billion and 5.3 billion triples, we obtain nearly linear speedup except for two portions: file I/O, which can be ameliorated with the additional service nodes, and data structure initialization, which requires nearly constant time for runs involving 32 processors or more.

99 GENERAL AND MISCELLANEOUS↗

A Framework for Incorporating General Domain Knowledge into Latent Dirichlet Allocation using First-Order Logic

Topic models have been used successfully for a variety of problems, often in the form of application-specific extensions of the basic Latent Dirichlet Allocation (LDA) model. Because deriving these new models in order to encode domain knowledge can be difficult and time-consuming, we propose the Fold-all model, which allows the user to specify general domain knowledge in First-Order Logic (FOL). However, combining topic modeling with FOL can result in inference problems beyond the capabilities of existing techniques. We have therefore developed a scalable inference technique using stochastic gradient descent which may also be useful to the Markov Logic Network (MLN) research community. Experiments demonstrate the expressive power of Fold-all, as well as the scalability of our proposed inference method.

97 MATHEMATICS AND COMPUTING↗

Lifetime Prediction for Degradation of Solar Mirrors using Step-Stress Accelerated Testing (Presentation)

This research is to illustrate the use of statistical inference techniques in order to quantify the uncertainty surrounding reliability estimates in a step-stress accelerated degradation testing (SSADT) scenario. SSADT can be used when a researcher is faced with a resource-constrained environment, e.g., limits on chamber time or on the number of units to test. We apply the SSADT methodology to adegradation experiment involving concentrated solar power (CSP) mirrors and compare the results to a more traditional multiple accelerated testing paradigm. Specifically, our work includes: (1) designing a durability testing plan for solar mirrors (3M's new improved silvered acrylic 'Solar Reflector Film (SFM) 1100') through the ultra-accelerated weathering system (UAWS), (2) definingdegradation paths of optical performance based on the SSADT model which is accelerated by high UV-radiant exposure, and (3) developing service lifetime prediction models for solar mirrors using advanced statistical inference. We use the method of least squares to estimate the model parameters and this serves as the basis for the statistical inference in SSADT. Several quantities of interest canbe estimated from this procedure, e.g., mean-time-to-failure (MTTF) and warranty time. The methods allow for the estimation of quantities that may be of interest to the domain scientists.

14 SOLAR ENERGY↗

Deep Spatiotemporal Feature Learning with Application to Image Classification

Deep machine learning is an emerging framework for dealing with complex high-dimensionality data in a hierarchical fashion which draws some inspiration from biological sources. Despite the notable progress made in the field, there remains a need for an architecture that can represent temporal information with the same ease that spatial information is discovered. In this work, we present new results using a recently introduced deep learning architecture called Deep Spatio-Temporal Inference Network (DeSTIN). DeSTIN is a discriminative deep learning architecture that combines concepts from unsupervised learning for dynamic pattern representation together with Bayesian inference. In DeSTIN the spatiotemporal dependencies that exist within the observations are modeled inherently in an unguided manner. Each node models the inputs by means of clustering and simple dynamics modeling while it constructs a belief state over the distribution of sequences using Bayesian inference. We demonstrate that information from the different layers of this hierarchical system can be extracted and utilized for the purpose of pattern classification. Earlier simulation results indicated that the framework is highly promising, consequently in this work we expand DeSTIN to a popular problem, the MNIST data set of handwritten digits. The system as a preprocessor to a neural network achieves a recognition accuracy of 97.98% on this data set. We further show related experimental results pertaining to automatic cluster adaptation and termination.

99 GENERAL AND MISCELLANEOUS↗

Uncertainty quantification given discontinuous climate model response and a limited number of model runs.

Uncertainty quantification in complex climate models is challenged by the sparsity of available climate model predictions due to the high computational cost of model runs. Another feature that prevents classical uncertainty analysis from being readily applicable is bifurcative behavior in climate model response with respect to certain input parameters. A typical example is the Atlantic Meridional Overturning Circulation. The predicted maximum overturning stream function exhibits discontinuity across a curve in the space of two uncertain parameters, namely climate sensitivity and CO2 forcing. We outline a methodology for uncertainty quantification given discontinuous model response and a limited number of model runs. Our approach is two-fold. First we detect the discontinuity with Bayesian inference, thus obtaining a probabilistic representation of the discontinuity curve shape and location for arbitrarily distributed input parameter values. Then, we construct spectral representations of uncertainty, using Polynomial Chaos (PC) expansions on either side of the discontinuity curve, leading to an averaged-PC representation of the forward model that allows efficient uncertainty quantification. The approach is enabled by a Rosenblatt transformation that maps each side of the discontinuity to regular domains where desirable orthogonality properties for the spectral bases hold. We obtain PC modes by either orthogonal projection or Bayesian inference, and argue for a hybrid approach that targets a balance between the accuracy provided by the orthogonal projection and the flexibility provided by the Bayesian inference - where the latter allows obtaining reasonable expansions without extra forward model runs. The model output, and its associated uncertainty at specific design points, are then computed by taking an ensemble average over PC expansions corresponding to possible realizations of the discontinuity curve. The methodology is tested on synthetic examples of discontinuous model data with adjustable sharpness and structure.

54 ENVIRONMENTAL SCIENCES↗

Learning planar ising models

Inference and learning of graphical models are both well-studied problems in statistics and machine learning that have found many applications in science and engineering. However, exact inference is intractable in general graphical models, which suggests the problem of seeking the best approximation to a collection of random variables within some tractable family of graphical models. In this paper, we focus our attention on the class of planar Ising models, for which inference is tractable using techniques of statistical physics [Kac and Ward; Kasteleyn]. Based on these techniques and recent methods for planarity testing and planar embedding [Chrobak and Payne], we propose a simple greedy algorithm for learning the best planar Ising model to approximate an arbitrary collection of binary random variables (possibly from sample data). Given the set of all pairwise correlations among variables, we select a planar graph and optimal planar Ising model defined on this graph to best approximate that set of correlations. We present the results of numerical experiments evaluating the performance of our algorithm.

36 MATERIALS SCIENCE↗

Comparison of Feynman Variance-To-Mean Measurements Using Measured and Simulated Data

Measured and simulated list-mode data from the NPOD detector system were compared for subcritical systems. Five BeRP ball configurations reflected by polyethylene are investigated in this work. The simulated data were calculated using the MCNP multiplication patch developed at LANL. In addition to comparing the moments of the counting distributions, a full analysis was conducted to infer the multiplication of each system using both the Hansen-Dowdy and Hage-Cifarelli formulisms. Several of the parameters that are input into these methods were also varied to determine their affect on the system multiplication. The main parameters investigated included the system source strength, reflector transmission, and the average number of neutrons released per 239 Pu fission ($\nu$). It is shown that when comparing the measured and simulated data while using the same method and parameters the inferred multiplication values agree within 5%. The measured and simulated multiplication values differ more as the reflector thickness increases. The moments for the simulated data were higher than the measured data for all of the configurations and also disagreed more as the reflector thickness increased. The analysis method and parameters used can lead to a large spread of inferred multiplication values. In the future more configurations will be analyzed in a similar manner and a detailed uncertainty analysis will be performed.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Learning planar Ising models

Inference and learning of graphical models are both well-studied problems in statistics and machine learning that have found many applications in science and engineering. However, exact inference is intractable in general graphical models, which suggests the problem of seeking the best approximation to a collection of random variables within some tractable family of graphical models. In this paper, we focus on the class of planar Ising models, for which exact inference is tractable using techniques of statistical physics. Based on these techniques and recent methods for planarity testing and planar embedding, we propose a greedy algorithm for learning the best planar Ising model to approximate an arbitrary collection of binary random variables (possibly from sample data). Given the set of all pairwise correlations among variables, we select a planar graph and optimal planar Ising model defined on this graph to best approximate that set of correlations. Finally, we demonstrate our method in simulations and for two applications: modeling senate voting records and identifying geo-chemical depth trends from Mars rover data.

97 MATHEMATICS AND COMPUTING↗

Robotic Understanding of Spatial Relationships Using Neural-Logic Learning

Understanding spatial relations of objects is critical in many robotic applications such as grasping, manipulation, and obstacle avoidance. Humans can simply reason object's spatial relations from a glimpse of a scene based on prior knowledge of spatial constraints. The proposed method enables a robot to comprehend spatial relationships among objects from RGB-D data. This paper proposed a neural-logic learning framework to learn and reason spatial relations from raw data by following logic rules on spatial constraints. The neural-logic network consists of three blocks: grounding block, spatial logic block, and inference block. The grounding block extracts high-level features from the raw sensory data. The spatial logic blocks can predicate fundamental spatial relations by training a neural network with spatial constraints. The inference block can infer complex spatial relations based on the predicated fundamental spatial relations. Simulations and robotic experiments evaluated the performance of the proposed method.

Wang, Dali↗

Machine Learning 5G Attack Detection in Programmable Logic

Machine learning-assisted network security may significantly contribute to securing 5G components. However, machine learning network security inference speeds generally require tens to hundreds of milliseconds thereby introducing significant latency in 5G operations. The inference latency can be reduced by deploying the machine learning model to programmable logic in a field programmable gate array (FPGA) at the cost of a small loss in accuracy. In order to quantify this loss, as well as to establish baseline performance inference speeds for programmable logic implementations, this work explores an autoencoder and a ß-variational autoencoder deployed on two different FPGA evaluation boards and compares accuracy and performance against an NVIDIA A100 GPU implementation. A publicly available 5G dataset containing 10 types of attacks along with normal traffic is introduced as part of the evaluation.

97 MATHEMATICS AND COMPUTING↗

Predicting Missing Regions in Charged Particle Tracks Using a Sparse 3D Convolutional Neural Network

The 2x2 Demonstrator is a prototype detector for the Deep Underground Neutrino Experiment (DUNE)'s Near Detector. Both the 2x2 Demonstrator and the Near Detector itself will have inactive regions wherein there is no sensitivity to charge deposition and light signals that arise from charged particle interactions with liquid argon. In the 2x2, these inactive regions are positioned in-between the active detector modules, which introduces the challenge of inferring what charge signals ought to look like in these regions. This study explores the use of a Sparse 3D Convolutional Neural Network (ConvNet) to infer missing regions in charged particle tracks. Hits corresponding to energy depositions are voxelized into a three-dimensional (3D) grid for each track. Inactive regions within the tracks are replaced with a dense, rectangular 3D grid of voxels, ensuring consistent step sizes in X, Y, and Z directions. Voxels in these dense regions are initialized with an energy value of -1, indicating nonphysical energy or charge. The model is trained to predict which voxels should activate as part of the track and which should not, with the goal of eventually inferring the missing charge or energy values in these voxels. Results indicate that the model accurately predicts track voxels within 1 unit in X, Y, or Z directions and effectively identifies non-track voxels, despite some overprediction. The approach shows promise in prediction of missing track regions with some accuracy.

Utaegbulam, Hilary↗

OPER: Optimality-Guided Embedding Table Parallelization for Large-scale Recommendation Model

With the sharp increasing volume of user data, Deep Learning Recommendation Model (DLRM) becomes an indispensable infrastructure in large technology companies. However, large-scale DLRM on the multi-GPU platform is still inefficient due to unbalanced workload partitioning and intensive inter-GPU communication. To this end, we propose OPER, an OPtimality guided Embedding table placement for large-scale Recommendation model training and inference. OPER explores the potential of mitigating remote memory access latency in DLRM through fine-grained embedding table placement. Specifically, OPER proposes a theoretical modeling that builds up the relationship between EMT placement and the embedding communication latency in both training and inference. OPER proves the NP hardness of finding the optimal embedding table placement and proposes a heuristic algorithm that yields near optimal placement. OPER implements a SHMEM-based embedding table training system and a unified embedding index mapping to support fine-grained embedding table sharding and placement. Comprehensive experiments reveal that OPER achieves on average 3.4× and 5.1× speedup on training and inference respectively over state-of-the-art DLRM frameworks.

Wang, Zheng↗

Nature-GL: A Revolutionary Learning Paradigm Unleashing Nature’s Power in Real-World Spatial-Temporal Graph Learning

Spatial-Temporal Graph Learning (ST-GL) is a prominent research area due to its unique capability to effectively learn real-world graphs. Applications of ST-GL pose stringent and various demands on not only real-time inference with low energy cost and high ac- curacy but also fast training. Unfortunately, as Moore’s Law approaches its limits and ST-GL model complexity drastically grows, the gap between digital hardware’s computational power and ST- GL application demands is widening. In response, this paper introduces Nature-GL, a nature-powered graph learning paradigm that exploits the principle of entropy increase to advance graph learning. In particular, Nature-GL transforms both the training and inference of real-valued ST-GL into electron-speed natural anneal- ing processes of a parameterized dynamical system that represents the target graphs. Experimental results across four real-world ap- plications with six datasets demonstrate that Nature-GL achieves orders-of-magnitude speedups in both training and inference, delivering higher accuracy compared to Graph Neural Networks.

Liu, Chuan [University of Rochester]↗

Lens Model Accuracy in the Expected LSST Lensed AGN Sample

Strong gravitational lensing of active galactic nuclei (AGN) enables measurements of cosmological parameters through time-delay cosmography (TDC). With data from the upcoming LSST survey, we anticipate using a sample of O(1000) lensed AGN for TDC. To prepare for this dataset and enable this measurement, we construct and analyze a realistic mock sample of 1300 systems drawn from the OM10 (Oguri & Marshall 2010) catalog of simulated lenses with AGN sources at $z<3.1$ in order to test a key aspect of the analysis pipeline, that of the lens modeling. We realize the lenses as power law elliptical mass distributions and simulate 5-year LSST i-band coadd images. From every image, we infer the lens mass model parameters using neural posterior estimation (NPE). Focusing on the key model parameters, $θ_E$ (the Einstein Radius) and $γ_{lens}$ (the projected mass density profile slope), with consistent mass-light ellipticity correlations in test and training data, we recover $θ_E$ with less than 1% bias per lens, 6.5% precision per lens and $γ_{lens}$ with less than 3% bias per lens, 8% precision per lens. We find that lens light subtraction prior to modeling is only useful when applied to data sampled from the training prior. If emulated deconvolution is applied to the data prior to modeling, precision improves across all parameters by a factor of 2. Finally, we combine the inferred lens mass models using Bayesian Hierarchical Inference to recover the global properties of the lens sample with less than 1% bias.

Venkatraman, Padmavathi [Illinois U., Urbana; KIPA↗

Strong Lensing Cosmology with Population-level Calibrated Neural Ratio Estimation

Strong gravitational lensing contains key information about cosmic acceleration. Modern and next-generation galaxy imaging surveys are expected to provide high-quality data on $\mathcal{O}(10^5)$ galaxy-galaxy lensing systems. The plethora and complexity of the data are likely to present computational challenges for parameter inference methods for fitting high-dimensional likelihoods, which are often analytically intractable. Neural Ratio Estimation (NRE) efficiently computes individual likelihood ratios that can be combined into population-level posteriors. We use simulations to study the capacity of NRE to jointly predict the dark energy equation-of-state parameter $w$ and the total matter density $Ω_{m}$ from lensing images and companion spectroscopic information. We also introduce a post hoc posterior coverage calibration procedure that mitigates the model overconfidence that is typically found in neural density estimation applications. Our experiments show that the errors on both parameters decrease with increasing inference population sizes. In particular, for 100 lenses in a standard $Λ$CDM Universe, our calibrated NRE model achieves median fractional uncertainty of $22.8\%$ in $w$ and $2.9\%$ in $Ω_{m}$. This proof of concept demonstrates a potentially scalable approach for efficient cosmological parameter inference with large populations of galaxy-scale lenses observed in future surveys.

Jarugula, Sreevani [Fermilab] (ORCID:0000000253867↗

Sensitivity of NIF-scale backlit thin shell implosions to hohlraum symmetry in the foot of the ignition drive pulse

A necessary condition for igniting indirectly-driven inertial confinement fusion (ICF) spherical capsules on the National Ignition Facility (NIF) is controlling drive flux asymmetry to the 1% level time-integrated over the pulse and with < 10%/ns swings during the pulse [J. D. Lindl et al., 'The Physics Basis for Ignition using Indirect Drive Targets on the National Ignition Facility', Physics of Plasmas 11, 339 (2003)]. While drive symmetry during the first 2 ns of the pulse can be inferred by using the re-emission pattern from a surrogate high Z sphere [E. Dewald et al. to be published in Rev. Sci. Inst.] and symmetry during the last 5 ns inferred from the shape of fully imploded capsules [A. Hauer, N. Delamater, D. Ress et al. Rev. Sci. Instrum. 66, 672-7 (1995)], the midportion ({approx} 2-10 ns) has been shown to be amenable to detection by the in-flight shape of x-ray backlit thin shell capsules [Pollaine et. al., Physics of Plasmas 8 2357 (2001)]. In this paper, we present sensitivity studies conducted on the University of Rochester's OMEGA laser of the thin shell symmetry measurement technique at near NIF-scale for two candidate capsule ablator materials, Ge-doped CH and Cu-doped Be. These experiments use both point and area backlighting to cast 4.7 keV radiographs of thin 1.4 mm initial-diameter Ge-doped CH and Cu-doped Be shells when converged a factor of {approx} 0.5 x in radius. Distortions in the position of the transmission limb of the shells resulting from drive asymmetries are measured to an accuracy of a few {micro}ms, meeting requirements. The promising results to date allow us to compare measured and predicted distortions and by inference drive asymmetries for the first 4 asymmetry modes as a function of hohlraum illumination conditions.

42 ENGINEERING↗

Non-intrusive reduced-order modeling for dynamical systems with spatially localized features

This work presents a non-intrusive reduced-order modeling framework for dynamical systems with spatially localized features characterized by slow singular value decay. The proposed approach builds upon two existing methodologies for reduced and full-order non-intrusive modeling, namely Operator Inference (OpInf) and sparse Full-Order Model (sFOM) inference. We decompose the domain into two complementary subdomains that exhibit fast and slow singular value decay. The dynamics of the subdomain exhibiting slow singular value decay are learned with sFOM while the dynamics with intrinsically low dimensionality on the complementary subdomain are learned with OpInf. The resulting, coupled OpInf-sFOM formulation leverages the computational efficiency of OpInf and the high resolution of sFOM, and thus enables fast non-intrusive predictions for conditions beyond those sampled in the training data set. A novel regularization technique with a closed-form solution based on the Gershgorin disk theorem is introduced to promote stable sFOM and OpInf models. We also provide a data-driven indicator for subdomain selection and ensure solution smoothness over the interface via a post-processing interpolation step. We evaluate the efficiency of the approach in terms of offline and online speedup through a quantitative, parametric computational cost analysis. We demonstrate the coupled OpInf-sFOM formulation for two test cases: a one-dimensional Burgers’ model for which accurate predictions beyond the span of the training snapshots are presented, and a two-dimensional parametric model for the Pine Island Glacier ice thickness dynamics, for which the OpInf-sFOM model achieves an average prediction error on the order of 1% with an online speedup factor of approximately 8$\times$ compared to the numerical simulation.

42 ENGINEERING↗

Approximate Bayesian Model Inversion for PDEs with Heterogeneous and State-Dependent Coefficients

We present two approximate Bayesian inference methods for parameter estimation in partial differential equation (PDE) models with space-dependent and state-dependent parameters. We demonstrate that these methods provide accurate and cost-effective alternatives to Markov Chain Monte Carlo simulation. We assume a parameterized Gaussian prior on the unknown functions, and approximate the posterior density by a parameterized multivariate Gaussian density. The parameters of the prior and posterior are estimated from sparse observations of the PDE model's states and the unknown functions themselves by maximizing the evidence lower bound (ELBO), a lower bound on the log marginal likelihood of the observations. The first method, Laplace-EM, employs the expectation maximization algorithm to maximize the ELBO, with a Laplace approximation of the posterior on the E-step, and minimization of a Kullback-Leibler divergence on the M-step. The second method, DSVI-EB, employs the doubly stochastic variational inference (DSVI) algorithm, in which the ELBO is maximized via gradient-based stochastic optimization, with nosiy gradients computed via simple Monte Carlo sampling and Gaussian backpropagation. We apply these methods to identifying diffusion coefficients in linear and nonlinear diffusion equations, and we find that both methods provide accurate estimates of posterior densities and the hyperparameters of Gaussian priors. While the Laplace-EM method is more accurate, it requires computing Hessians of the physics model. The DSVI-EB method is found to be less accurate but only requires gradients of the physics model.

approximate Bayesian inference, model inversion, v↗