Search NASA⌕ Search

SEARCH · Search NASA

Results for “Loss function”

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 181 records · Page 10

Hutchinson Trace Estimation for high-dimensional and high-order Physics-Informed Neural Networks

Physics-Informed Neural Networks (PINNs) have proven effective in solving partial differential equations (PDEs), especially when some data are available by seamlessly blending data and physics. However, extending PINNs to high-dimensional and even high-order PDEs encounters significant challenges due to the computational cost associated with automatic differentiation in the residual loss function calculation. Herein, we address the limitations of PINNs in handling high-dimensional and high-order PDEs by introducing the Hutchinson Trace Estimation (HTE) method. Starting with the second-order high-dimensional PDEs, which are ubiquitous in scientific computing, HTE is applied to transform the calculation of the entire Hessian matrix into a Hessian vector product (HVP). This approach not only alleviates the computational bottleneck via Taylor-mode automatic differentiation but also significantly reduces memory consumption from the Hessian matrix to an HVP’s scalar output. We further showcase HTE’s convergence to the original PINN loss and its unbiased behavior under specific conditions. Comparisons with the Stochastic Dimension Gradient Descent (SDGD) highlight the distinct advantages of HTE, particularly in scenarios with significant variability and variance among dimensions. We further extend the application of HTE to higher-order and higher-dimensional PDEs, specifically addressing the biharmonic equation. By employing tensor-vector products (TVP), HTE efficiently computes the colossal tensor associated with the fourth-order high-dimensional biharmonic equation, saving memory and enabling rapid computation. The effectiveness of HTE is illustrated through experimental setups, demonstrating comparable convergence rates with SDGD under memory and speed constraints. Additionally, HTE proves valuable in accelerating the Gradient-Enhanced PINN (gPINN) version as well as the Biharmonic equation. Overall, HTE opens up a new capability in scientific machine learning for tackling high-order and high-dimensional PDEs.

Curse of dimensionality↗

Differentiable programming for online training of a neural artificial viscosity function within a staggered grid Lagrangian hydrodynamics scheme

Lagrangian methods to solve the inviscid Euler equations produce numerical oscillations near shock waves. A common approach to reducing these oscillations is to add artificial viscosity (AV) to the discrete equations. The AV term acts as a dissipative mechanism that attenuates oscillations by smearing the shock across a finite number of computational cells. However, AV introduces several control parameters that are not determined by the underlying physical model, and hence, in practice are tuned to the characteristics of a given problem. We seek to improve the standard quadratic-linear AV form by replacing it with a learned neural function that reduces oscillations relative to exact solutions of the Euler equations, resulting in a hybrid numerical-neural hydrodynamic solver. Because AV is an artificial construct that exists solely to improve the numerical properties of a hydrodynamic code, there is no offline ‘viscosity data’ against which a neural network can be trained before inserting into a numerical simulation, thus requiring online training. We achieve this via differentiable programming, i.e. end-to-end backpropagation or adjoint solution through both the neural and differential equation code, using automatic differentiation of the hybrid code in the Julia programming language to calculate the necessary loss function gradients. A novel offline pre-training step accelerates training by initializing the neural network to the default numerical AV scheme, which can be learned rapidly by space-filling sampling over the AV input space. We find that online training over early time steps of simulation is sufficient to learn a neural AV function that reduces numerical oscillations in long-term hydrodynamic shock simulations. These results offer an early proof-of-principle that online differentiable training of hybrid numerical schemes with novel neural network components can improve certain performance aspects existing in purely numerical schemes.

97 MATHEMATICS AND COMPUTING↗

Solving Inverse Stochastic Problems from Discrete Particle Observations Using the Fokker--Planck Equation and Physics-Informed Neural Networks

The Fokker--Planck (FP) equation governing the evolution of the probability density function (PDF) is applicable to many disciplines, but it requires specification of the coefficients for each case, which can be functions of space-time and not just constants and hence require the development of a data-driven modeling approach. When the data available is directly on the PDF, there exist methods for inverse problems that can be employed to infer the coefficients and thus determine the FP equation and subsequently obtain its solution. Herein, we address a more realistic scenario, where only sparse data are given on the particles' positions at a few time instants, which are not sufficient to accurately construct directly the PDF even at those times from existing methods, e.g., kernel estimation algorithms. To this end, we develop a general framework based on physics-informed neural networks (PINNs) that introduces a new loss function using the Kullback--Leibler divergence to connect the stochastic samples with the FP equation to simultaneously learn the equation and infer the multidimensional PDF at all times. In particular, we consider two types of inverse problems, type I, where the FP equation is known but the initial PDF is unknown, and type II, in which, in addition to the unknown initial PDF, the drift and diffusion terms are also unknown. In both cases, we investigate problems with either Brownian or Lévy noise or a combination of both. Here, we demonstrate the new PINN framework in detail in the one-dimensional (1D) case, but we also provide results for up to five dimensions demonstrating that we can infer both the FP equation and dynamics simultaneously at all times with high accuracy using only very few discrete observations of the particles.

97 MATHEMATICS AND COMPUTING↗

SO(3)-invariance of informed-graph-based deep neural network for anisotropic elastoplastic materials

This work examines the frame-invariance (and the lack thereof) exhibited in simulated anisotropic elasto-plastic responses generated from supervised machine learning of classical multi-layer and informed-graph-based neural networks, and proposes different remedies to fix this drawback. The inherent hierarchical relations among physical quantities and state variables in an elasto-plasticity model are first represented as informed, directed graphs, where three variations of the graph are tested. While feed-forward neural networks are used to train path-independent constitutive relations (e.g., elasticity), recurrent neural networks are used to replicate responses that depends on the deformation history, i.e. or path dependent. In dealing with the objectivity deficiency, we use the spectral form to represent tensors and, subsequently, three metrics, the Euclidean distance between the Euler Angles, the distance from the identity matrix, and geodesic on the unit sphere in Lie algebra, can be employed to constitute objective functions for the supervised machine learning. In this, the aim is to minimize the measured distance between the true and the predicted 3D rotation entities. Following this, we conduct numerical experiments on how these metrics, which are theoretically equivalent, may lead to differences in the efficiency of the supervised machine learning as well as the accuracy and robustness of the resultant models. Neural network models trained with tensors represented in component form for a given Cartesian coordinate system are used as a benchmark. Our numerical tests show that, even given the same amount of information and data, the quality of the anisotropic elasto-plasticity model is highly sensitive to the way tensors are represented and measured. The results reveal that using a loss function based on geodesic on the unit sphere in Lie algebra together with an informed, directed graph yield significantly more accurate rotation prediction than the other tested approaches.

42 ENGINEERING↗

Robust scalable initialization for Bayesian variational inference with multi-modal Laplace approximations

Predictive modeling typically relies on Bayesian model calibration to provide uncertainty quantification. Variational inference utilizing fully independent (“mean-field”) Gaussian distributions are often used as approximate probability density functions. This simplification is attractive since the number of variational parameters grows only linearly with the number of unknown model parameters. However, the resulting diagonal covariance structure and unimodal behavior can be too restrictive to provide useful approximations of intractable Bayesian posteriors that exhibit highly non-Gaussian behavior, including multimodality. High-fidelity surrogate posteriors for these problems can be obtained by considering the family of Gaussian mixtures. Gaussian mixtures are capable of capturing multiple modes and approximating any distribution to an arbitrary degree of accuracy, while maintaining some analytical tractability. Unfortunately, variational inference using Gaussian mixtures with full-covariance structures suffers from a quadratic growth in variational parameters with the number of model parameters. The existence of multiple local minima due to strong nonconvex trends in the loss functions often associated with variational inference present additional complications, These challenges motivate the need for robust initialization procedures to improve the performance and computational scalability of variational inference with mixture models. In this work, we propose a method for constructing an initial Gaussian mixture model approximation that can be used to warm-start the iterative solvers for variational inference. The procedure begins with a global optimization stage in model parameter space. In this step, local gradient-based optimization, globalized through multistart, is used to determine a set of local maxima, which we take to approximate the mixture component centers. Around each mode, a local Gaussian approximation is constructed via the Laplace approximation. Finally, the mixture weights are determined through constrained least squares regression. The robustness and scalability of the proposed methodology is demonstrated through application to an ensemble of synthetic tests using high-dimensional, multimodal probability density functions. Here, the practical aspects of the approach are demonstrated with inversion problems in structural dynamics.

97 MATHEMATICS AND COMPUTING↗

QUASI-PERIODIC FLUCTUATIONS AND CHROMOSPHERIC EVAPORATION IN A SOLAR FLARE RIBBON OBSERVED BY HINODE /EIS, IRIS , AND RHESSI

The Hinode /Extreme-ultraviolet Imaging Spectrometer (EIS) obtained rapid cadence (11.2 s) EUV stare spectra of an M7.3 flare ribbon in AR 12036 on 2014 April 18. Quasi-periodic ( P ≈ 75.6 ± 9.2 s) intensity fluctuations occurred in emission lines of O iv, Mg vi, Mg vii, Si vii, Fe xiv, and Fe xvi during the flare's impulsive rise, and ended when the maximum intensity in Fe xxiii was reached. The profiles of the O iv–Fe xvi lines reveal that they were all redshifted during most of the interval of quasi-periodic intensity fluctuations, while the Fe xxiii profile revealed multiple components including one or two highly blueshifted ones. This indicates that the flare underwent explosive chromospheric evaporation during its impulsive rise. Fluctuations in the relative Doppler velocities were seen, but their amplitudes were too subtle to extract significant quasi-periodicities. RHESSI detected 25–100 keV hard-X-ray sources in the ribbon near the EIS slit's pointing position during the peaks in the EIS intensity fluctuations. The observations are consistent with a series of energy injections into the chromosphere by nonthermal particle beams. Electron densities derived with Fe xiv (4.6 × 10{sup 10} cm{sup −3}) and Mg vii (7.8 × 10{sup 9} cm{sup −3}) average line intensity ratios during the interval of quasi-periodic intensity fluctuations, combined with the radiative loss function of an optically thin plasma, yield radiative cooling times of 32 s at 2.0 × 10{sup 6} K, and 46 s at 6.3 × 10{sup 5} K (about half the quasi-period); assuming Fe xiv's density for Fe xxiii yields a radiative cooling time of 10{sup 3} s (13 times the quasi-period) at 1.4 × 10{sup 7} K.

79 ASTRONOMY AND ASTROPHYSICS↗

Senescence-Associated Secretory Phenotypes Reveal Cell-Nonautonomous Functions of Oncogenic RAS and the p53 Tumor Suppressor

Cellular senescence suppresses cancer by arresting cell proliferation, essentially permanently, in response to oncogenic stimuli, including genotoxic stress. We modified the use of antibody arrays to provide a quantitative assessment of factors secreted by senescent cells. We show that human cells induced to senesce by genotoxic stress secrete myriad factors associated with inflammation and malignancy. This senescence-associated secretory phenotype (SASP) developed slowly over several days and only after DNA damage of sufficient magnitude to induce senescence. Remarkably similar SASPs developed in normal fibroblasts, normal epithelial cells, and epithelial tumor cells after genotoxic stress in culture, and in epithelial tumor cells in vivo after treatment of prostate cancer patients with DNA-damaging chemotherapy. In cultured premalignant epithelial cells, SASPs induced an epithelial-mesenchyme transition and invasiveness, hallmarks of malignancy, by a paracrine mechanism that depended largely on the SASP factors interleukin (IL)-6 and IL-8. Strikingly, two manipulations markedly amplified, and accelerated development of, the SASPs: oncogenic RAS expression, which causes genotoxic stress and senescence in normal cells, and functional loss of the p53 tumor suppressor protein. Both loss of p53 and gain of oncogenic RAS also exacerbated the promalignant paracrine activities of the SASPs. Our findings define a central feature of genotoxic stress-induced senescence. Moreover, they suggest a cell-nonautonomous mechanism by which p53 can restrain, and oncogenic RAS can promote, the development of age-related cancer by altering the tissue microenvironment.

59 BASIC BIOLOGICAL SCIENCES↗

Layer specific optical band gap measurement at nanoscale in MoS{sub 2} and ReS{sub 2} van der Waals compounds by high resolution electron energy loss spectroscopy

Layer specific direct measurement of optical band gaps of two important van der Waals compounds, MoS{sub 2} and ReS{sub 2}, is performed at nanoscale by high resolution electron energy loss spectroscopy. For monolayer MoS{sub 2}, the twin excitons (1.8 and 1.95 eV) originating at the K point of the Brillouin zone are observed. An indirect band gap of 1.27 eV is obtained from the multilayer regions. Indirect to direct band gap crossover is observed which is consistent with the previously reported strong photoluminescence from the monolayer MoS{sub 2}. For ReS{sub 2}, the band gap is direct, and a value of 1.52 and 1.42 eV is obtained for the monolayer and multilayer, respectively. The energy loss function is dominated by features due to high density of states at both the valence and conduction band edges, and the difference in analyzing band gap with respect to ZnO is highlighted. Crystalline 1T ReS{sub 2} forms two dimensional chains like superstructure due to the clustering between four Re atoms. The results demonstrate the power of HREELS technique as a nanoscale optical absorption spectroscopy tool.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Analytic expressions for the inelastic scattering and energy loss of electron and proton beams in carbon nanotubes

We have determined ''effective'' Bethe coefficients and the mean excitation energy of stopping theory (I-value) for multiwalled carbon nanotubes (MWCNTs) and single-walled carbon nanotube (SWCNT) bundles based on a sum-rule constrained optical-data model energy loss function with improved asymptotic properties. Noticeable differences between MWCNTs, SWCNT bundles, and the three allotropes of carbon (diamond, graphite, glassy carbon) are found. By means of Bethe's asymptotic approximation, the inelastic scattering cross section, the electronic stopping power, and the average energy transfer to target electrons in a single inelastic collision, are calculated analytically for a broad range of electron and proton beam energies using realistic excitation parameters.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

EFFECT OF FINITE LARMOR RADIUS CORRECTIONS ON THE THERMAL INSTABILITY OF THERMALLY CONDUCTING VISCOUS PLASMA WITH HALL CURRENT AND ELECTRON INERTIA

The thermal instability of an infinite homogeneous, thermally conducting, and rotating plasma, incorporating finite electrical resistivity, finite electron inertia, and an arbitrary radiative heat-loss function in the presence of finite Larmor radius corrections and Hall current, has been studied. Analysis has been made with the help of linearized magnetohydrodynamics (MHD) equations. A general dispersion relation is obtained using the normal mode analysis method, and the dispersion relation is discussed for longitudinal propagation and transverse propagation separately. The dispersion relation has been solved numerically to obtain the dependence of the growth rate on the various parameters involved. The conditions of modified thermal instability and stability are discussed in the different cases of interest.

79 ASTRONOMY AND ASTROPHYSICS↗

Bandgap engineering of SrZrS 3 chalcogenide perovskite via substitutional doping for photovoltaic applications: a first-principles DFT study

Strontium zirconium sulfide (SrZrS 3 ) has garnered significant attention for photovoltaic (PV) applications due to its excellent optoelectronic properties, high chemical and moisture stability, and non-toxicity. However, the bandgaps of both the α- and β-phases lie outside the optimum ranges for both single-junction solar cells (SJSCs) and tandem solar cells (TSCs), thereby limiting their applications in PV technologies. In this study, we employed hybrid density functional theory to engineer the band gaps of α- and β-SrZrS 3 through substitutional doping. We considered three dopants (Hf, Sn, and Ti) at the Zr site of SrZrS 3 with varying doping concentrations and investigated their effects on the structural, electronic, and optical properties of the materials. We found that Sn and Ti doping effectively lowers the band gaps of both α- and β-SrZrS 3 , whereas Hf doping increases them. For x values up to 0.25, the band gaps of the α-SrZr 1−x Sn x S 3 and α-SrZr 1−x Ti x S 3 are within the optimum range for SJSCs, and those of β-SrZr 1−x Sn x S 3 and β-SrZr 1−x Ti x S 3 lie within the optimum range for Si/perovskite as well as perovskite/perovskite TSCs. The three dopants exhibited significant effects on the optical properties of both α- and β-SrZrS 3 , including the absorption coefficient, energy-loss functions, reflectivity, and refractivity spectra. Thermodynamic stability analysis revealed that for both phases, SrZr 1−x Hf x S 3 can be synthesized via exothermic processes, whereas the formation of SrZr 1−x Ti x S 3 and SrZr1−xSn x S 3 is endothermic and hence, not thermodynamically favorable. Further analysis showed that SrZr 1−x Ti x S 3 in both α- and β-phases are stable under thermodynamic equilibrium conditions, whereas SrZr 1−x Sn x S 3 is prone to dissociation into ternary phases (SrZrS 3 and SrSnS 3 ), especially at higher doping concentrations. These results show that Ti doping is effective in tuning the band gaps of α- and β-SrZrS 3 toward the optimal values for PV applications.

14 SOLAR ENERGY↗

Analyzing and Exploring Training Recipes for Large-Scale Transformer-Based Weather Prediction

Abstract The rapid rise of deep learning (DL) in numerical weather prediction (NWP) has led to a proliferation of models which forecast atmospheric variables with comparable or superior skill than traditional physics-based NWP. However, among these leading DL models, there is a wide variance in both the training settings and architecture used. Further, the lack of thorough ablation studies makes it hard to discern which components are most critical to success. In this work, we show that it is possible to attain high forecast skill even with relatively off-the-shelf architectures, simple training procedures, and moderate compute budgets. Specifically, we train a minimally modified Swin Transformer V2 (SwinV2) on ERA5 data and find that it attains superior skill in terms of mean-square errors of deterministic forecasts when compared against the European Centre for Medium-Range Weather Forecasts’ Integrated Forecasting System (IFS). Almost all DL–NWP systems share a core set of hyperparameters and design decisions. To aid and expedite future DL–NWP research, we present an in-depth, systematic exploration of different loss functions, model sizes and depths, patch sizes, and multistep training objectives. We also examine the model performance with metrics beyond the typical accuracy (ACC) and RMSE and investigate how the performance scales with model size. Through our open-source code, scoring pipelines, and models, we share our findings on key aspects of the training pipeline. These ablations reduce the necessity for expensive hyperparameter tuning and lower the barrier to entry for future DL–NWP research. Significance Statement This study investigates the potential of using large-scale transformer-based models for weather prediction, showing that it is possible to achieve high forecast accuracy with simpler, off-the-shelf architectures. By training a minimally modified SwinV2 transformer on ERA5 data, we show that the model achieves competitive forecast skill in terms of mean-square error for key variables, outperforming the European Centre for Medium-Range Weather Forecasts’ Integrated Forecasting System (IFS) at all lead times. Our findings suggest that effective training strategies, such as multistep fine-tuning and channel-weighted losses, significantly enhance the model’s performance. However, we also highlight that these improvements come with trade-offs in other areas, such as ensemble spread and high-frequency spatial detail. This work highlights the promise of deep learning in improving weather forecasts, which could lead to better preparedness and response to weather events, ultimately benefiting society by providing more reliable weather predictions.

Willard, Jared D. [Lawrence Berkeley National Labo↗

Ab-initio study of the structural, linear and nonlinear optical properties of CdAl{sub 2}Se{sub 4} defect-chalcopyrite

The complex density functional theory (DFT) calculations of structural, electronic, linear and nonlinear optical properties for the defect chalcopyrite CdAl{sub 2}Se{sub 4} compound have been reported using the full potential linearized augmented plane wave (FP-LAPW) method as implemented in the WIEN2k code. We employed the Wu and Cohen generalized gradient approximation (GGA-WC), which is based on exchange-correlation energy optimization to calculate the total energy. Also we have used the Engel-Vosko GGA formalism, which optimizes the corresponding potential for band structure, density of states and the spectral features of the linear and nonlinear optical properties. This compound has a wide direct energy band gap of about 2.927 eV with both the valence band maximum and conduction band minimum located at the center of the Brillouin zone. The ground state quantities such as lattice parameters (a, c, x, y and z), bulk modulus B and its pressure derivative B' are evaluated. We have calculated the frequency-dependent complex epsilon(omega), its zero-frequency limit epsilon{sub 1}(0), refractive index n(omega), birefringence DELTAn(omega), the reflectivity R(omega) and electron energy loss function L(omega). Calculations are reported for the frequency-dependent complex second-order nonlinear optical susceptibilities. We find opposite signs of the contributions of the 2omega and 1omega inter/intra-band to the imaginary part for the dominant component through the wide optical frequency range. - Graphical abstract: Calculated band structure and total density of CdAl{sub 2}Se{sub 4}.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Energy loss of proton, alpha particle, and electron beams in hafnium dioxide films

The electronic stopping power, S, of HfO{sub 2} films for proton and alpha particle beams has been measured and calculated. The experimental data have been obtained by the Rutherford backscattering technique and cover the range of 120-900 and 120-3000 keV for proton and alpha particle beams, respectively. Theoretical calculations of the energy loss for the same projectiles have been done by means of the dielectric formalism using the Mermin energy loss function--generalized oscillator strength (MELF-GOS) model for a proper description of the HfO{sub 2} target on the whole momentum-energy excitation spectrum. At low projectile energies, a nonlinear theory based on the extended Friedel sum rule has been employed. The calculations and experimental measurements show good agreement for protons and a quite good one for alpha particles. In particular, the experimental maximums of both stopping curves (around 120 and 800 keV, respectively) are well reproduced. On the basis of this good agreement, we have also calculated the inelastic mean-free path (IMFP) and the stopping power for electrons in HfO{sub 2} films. Our results predict a minimum value of the IMFP and a maximum value of the S for electrons with energies around 120 and 190 eV, respectively.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Linking age, survival, and transit time distributions

Abstract Although the concepts of age, survival, and transit time have been widely used in many fields, including population dynamics, chemical engineering, and hydrology, a comprehensive mathematical framework is still missing. Here we discuss several relationships among these quantities by starting from the evolution equation for the joint distribution of age and survival, from which the equations for age and survival time readily follow. It also becomes apparent how the statistical dependence between age and survival is directly related to either the age dependence of the loss function or the survival‐time dependence of the input function. The solution of the joint distribution equation also allows us to obtain the relationships between the age at exit (or death) and the survival time at input (or birth), as well as to stress the symmetries of the various distributions under time reversal. The transit time is then obtained as a sum of the age and survival time, and its properties are discussed along with the general relationships between their mean values. The special case of steady state case is analyzed in detail. Some examples, inspired by hydrologic applications, are presented to illustrate the theory with the specific results.

Calabrese, Salvatore↗

Accurate and Data‐Efficient Micro X‐ray Diffraction Phase Identification Using Multitask Learning: Application to Hydrothermal Fluids

Traditional analysis of highly distorted micro X‐ray diffraction (μ‐XRD) patterns from hydrothermal fluid environments is a time‐consuming process, often requiring substantial data preprocessing and labeled experimental data. Herein, the potential of deep learning with a multitask learning (MTL) architecture to overcome these limitations is demonstrated. MTL models are trained to identify phase information in μ‐XRD patterns, minimizing the need for labeled experimental data and masking preprocessing steps. Notably, MTL models show superior accuracy compared to binary classification convolutional neural networks. Additionally, introducing a tailored cross‐entropy loss function improves MTL model performance. Most significantly, MTL models tuned to analyze raw and unmasked XRD patterns achieve close performance to models analyzing preprocessed data, with minimal accuracy differences. This work indicates that advanced deep learning architectures like MTL can automate arduous data handling tasks, streamline the analysis of distorted XRD patterns, and reduce the reliance on labor‐intensive experimental datasets.

97 MATHEMATICS AND COMPUTING↗

Data‐driven variational method for discrepancy modeling: Dynamics with small‐strain nonlinear elasticity and viscoelasticity

Abstract The effective inclusion of a priori knowledge when embedding known data in physics‐based models of dynamical systems can ensure that the reconstructed model respects physical principles, while simultaneously improving the accuracy of the solution in the previously unseen regions of state space. This paper presents a physics‐constrained data‐driven discrepancy modeling method that variationally embeds known data in the modeling framework. The hierarchical structure of the method yields fine scale variational equations that facilitate the derivation of residuals which are comprised of the first‐principles theory and sensor‐based data from the dynamical system. The embedding of the sensor data via residual terms leads to discrepancy‐informed closure models that yield a method which is driven not only by boundary and initial conditions, but also by measurements that are taken at only a few observation points in the target system. Specifically, the data‐embedding term serves as residual‐based least‐squares loss function, thus retaining variational consistency. Another important relation arises from the interpretation of the stabilization tensor as a kernel function, thereby incorporating a priori knowledge of the problem and adding computational intelligence to the modeling framework. Numerical test cases show that when known data is taken into account, the data driven variational (DDV) method can correctly predict the system response in the presence of several types of discrepancies. Specifically, the damped solution and correct energy time histories are recovered by including known data in the undamped situation. Morlet wavelet analyses reveal that the surrogate problem with embedded data recovers the fundamental frequency band of the target system. The enhanced stability and accuracy of the DDV method is manifested via reconstructed displacement and velocity fields that yield time histories of strain and kinetic energies which match the target systems. The proposed DDV method also serves as a procedure for restoring eigenvalues and eigenvectors of a deficient dynamical system when known data is taken into account, as shown in the numerical test cases presented here.

Masud, Arif↗