Search NASA⌕ Search

SEARCH · Search NASA

Results for “higher-order networks”

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 19 records
All sourcesNASA NTRSDOE OSTI
NASA NTRS1991

Higher-order neural network software for distortion invariant object recognition

The state-of-the-art in pattern recognition for such applications as automatic target recognition and industrial robotic vision relies on digital image processing. We present a higher-order neural network model and software which performs the complete feature extraction-pattern classification paradigm required for automatic pattern recognition. Using a third-order neural network, we demonstrate complete, 100 percent accurate invariance to distortions of scale, position, and in-plate rotation. In a higher-order neural network, feature extraction is built into the network, and does not have to be learned. Only the relatively simple classification step must be learned. This is key to achieving very rapid training. The training set is much smaller than with standard neural network software because the higher-order network only has to be shown one view of each object to be learned, not every possible view. The software and graphical user interface run on any Sun workstation. Results of the use of the neural software in autonomous robotic vision systems are presented. Such a system could have extensive application in robotic manufacturing.

Reid, Max B.↗
NASA NTRS1990

Connectivity strategies for higher-order neural networks applied to pattern recognition

Different strategies for non-fully connected HONNs (higher-order neural networks) are discussed, showing that by using such strategies an input field of 128 x 128 pixels can be attained while still achieving in-plane rotation and translation-invariant recognition. These techniques allow HONNs to be used with the larger input scenes required for practical pattern-recognition applications. The number of interconnections that must be stored has been reduced by a factor of approximately 200,000 in a T/C case and about 2000 in a Space Shuttle/F-18 case by using regional connectivity. Third-order networks have been simulated using several connection strategies. The method found to work best is regional connectivity. The main advantages of this strategy are the following: (1) it considers features of various scales within the image and thus gets a better sample of what the image looks like; (2) it is invariant to shape-preserving geometric transformations, such as translation and rotation; (3) the connections are predetermined so that no extra computations are necessary during run time; and (4) it does not require any extra storage for recording which connections were formed.

Spirkovska, Lilly↗
DOE OSTI2026

TANTE: Time-adaptive operator learning via neural Taylor expansion

Operator learning for time-dependent partial differential equations (PDEs) has seen rapid progress in recent years, enabling efficient approximation of complex spatiotemporal dynamics. However, most existing methods rely on fixed time step sizes during rollout, which limits their ability to adapt to varying temporal complexity and often leads to error accumulation. In this work, we propose the Time-Adaptive Transformer with Neural Taylor Expansion (TANTE), a novel operator-learning framework that produces continuous-time predictions with adaptive step sizes. TANTE predicts future states by performing a Taylor expansion at the current state, where neural networks learn both the higher-order temporal derivatives and the local radius of convergence. This allows the model to dynamically adjust its rollout based on the local behavior of the solution, thereby reducing cumulative error and improving computational efficiency. We demonstrate the effectiveness of TANTE across a wide range of PDE benchmarks, achieving superior accuracy and adaptability compared to fixed-step baselines, delivering accuracy gains of 60-80 % and speed-ups of 30-40 % at inference time.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1983

The use of Ada in distributed simulations

The increasing need for detailed information about systems of continually growing complexity enhances steadily the demands regarding the employed models. The present investigation is concerned with work related to the development of high-performance computer hardware intended for the support of the real-time simulation of jet engines. The hardware is structured in the form of a network of communicating microprocessors running in parallel. The need for a higher-order language capability for programming such a network has led to the research considered in this study. Attention is given to the hardware which is being developed, an abstract model, programming language considerations, research considerations, research objectives, Ada tasks, Ada packages, the Ada model, the mapping of the model to the hardware, a precompiler example, and the advantages of Ada.

Collins, W. R.↗
DOE OSTI2025

Nanotubes Growth by Self-Assembly of DNA Strands at Room Temperature

Artificial biomolecular nanotubes are a promising approach to building materials mimicking the capacity of the cellular cytoskeleton to grow and self-organize dynamically. Nucleic acid nanotechnology has demonstrated a variety of self-assembling nanotubes with programmable, robust features and morphological similarities to actual cytoskeleton components. However, their production typically requires thermal annealing, which not only poses a general constraint on their potential applications but is also incompatible with physiological conditions. Here, we demonstrate that DNA nanotubes can self-assemble from a simple mixture of five short DNA strands at constant room temperature, growing for extended periods of time in bulk conditions as well as under confinement. Assembly is achieved using a monovalent salt buffer, which ensures a faithful nanoscale arrangement and avoids nanotube aggregation. We observe the formation of individual nanotubes up to 20 days with a diameter of 22 ± 4 nm and length of several tens of micrometers. We finally encapsulate the strands in microsized compartments, such as water-in-oil microdroplets and giant unilamellar vesicles serving as simple cell models. Notably, nanotubes not only isothermally self-assemble directly inside the microcompartments but also self-organize into dynamic higher-order structures resembling rings and dynamic networks. Our study provides an advantageous method for in situ assembly of programmable biomolecular scaffolds and materials using synthetic DNA strands without requirements of thermal treatment.

Chemistry↗
NASA NTRS1987

Analysis of a Four-Reflector S/X-Band Antenna

Physical optics accounts for near field, cross polarization, and higher-order modes. Report presents physical-optics analysis of four-reflector, 64-m antennas of Deep Space Network. Analysis thorough and detailed. Report has instructional value as example for designers of large microwave dishes with subreflectors and involving reflector surfaces with hyperboloidal, paraboloidal, ellipsoidal, and more complex shapes.

Cha, Alan G.↗
DOE OSTI2025

Direct observation of key aluminum hydroxide prenucleation oligomers for gibbsite nucleation and crystallization in sodium aluminate solution by liquid ToF-SIMS

The mechanism of gibbsite (aluminum hydroxide) crystallization from highly alkaline solutions such as Bayer liquors remains poorly understood, where aluminum (Al) transforms from largely tetrahedrally coordinated aluminate monomers in sodium aluminate solutions into a network of octahedra in gibbsite crystals. A variety of traditional analytical approaches applied to this system do not readily reveal the presence of higher-order oligomeric intermediates. To overcome this limitation, we employed in-situ liquid Time-of-Flight Secondary Ion Mass Spectrometry (ToF-SIMS) to examine the Al species present in concentrated sodium aluminate solutions favoring crystallization of gibbsite or sodium aluminate. A complex mixture of Al oligomers was found. By comparing the change in the relative concentration of Al oligomers with +1 and -1 charge, we were able to identify three major Al oligomer candidates, iso-tetramers, iso-pentamers, and cyclic-hexamers, for the nucleation and crystallization of gibbsite. The concentrations of iso-tetramers and iso-pentamers significantly surpass those of cyclic-hexamers. Time-dependent in-situ Raman spectroscopy analysis indicated that the appearance of gibbsite coincided with the peak concentration of these oligomers. The Density-functional theory (DFT) calculation suggests that the formation of iso-oligomers is more favorable than that of cyclic-hexamers. The combined results suggest that iso-tetramers and iso-pentamers play the most substantial role in the nucleation and growth of gibbsite in the sodium aluminate solutions. Our findings also suggest that the oligomers that promote gibbsite crystallization are more stable in dilute sodium aluminate solutions, making these solutions particularly suitable for efficient gibbsite crystallization. In conclusion, our study fills a major knowledge gap in understanding Al speciation that leads to the nucleation and crystallization of gibbsite in concentrated sodium aluminate solutions.

Bayer liquor↗
DOE OSTI2025

Gradient flow based phase-field modeling using separable neural networks

Allen–Cahn equation is a reaction–diffusion equation and is widely used for modeling phase separation. Machine learning methods for solving the Allen–Cahn equation in its strong form suffer from inaccuracies in collocation techniques, errors in computing higher-order spatial derivatives, and the large system size required by the space–time approach. To overcome these challenges, we propose solving the gradient flow of the Ginzburg–Landau free energy functional, which is equivalent to the Allen–Cahn equation, thereby avoiding the second-order spatial derivatives associated with the Allen–Cahn equation. A minimizing movement scheme is employed to solve the gradient flow problem, eliminating the complexities of a space–time approach. We utilize a separable neural network that efficiently represents the phase field through low-rank tensor decomposition. As we use the minimizing movement scheme to numerically solve the gradient flow problem, we thus, refer to the proposed method as the Separable Deep Minimizing Movement (SDMM) method. The evaluation of the functional in the minimizing movement scheme using the Gauss quadrature technique bypasses the inaccuracies associated with collocation techniques traditionally used to solve partial differential equations. A hyperbolic tangent transformation is introduced on the phase field prior to the evaluation of the functional to ensure that it remains strictly bounded within the values of the two phases. For this transformation, theoretical guarantee for energy stability of the minimizing movement scheme is established. Our results suggest that this transformation helps to improve the accuracy and efficiency significantly. The proposed method resolves the challenges faced by state-of-the-art machine learning techniques, outperforming them in both accuracy and efficiency. It is also the first machine learning method to achieve an order of magnitude speed improvement over the finite element method. In addition to its formulation and computational implementation, several case studies illustrate the applicability of the proposed method.

42 ENGINEERING↗
DOE OSTI2025

Ginsparg-Wilson Hamiltonians with Improved Chiral Symmetry

We construct a family of Ginsparg-Wilson Hamiltonians with improved chiral properties, starting from a construction of Creutz-Horvath-Neuberger that provides a doubler-free Hamiltonian lattice regularization for Dirac fermions in even spacetime dimensions. We use a higher-order generalization of the Ginsparg-Wilson relation due to Fujikawa, which yields an order-$k$ Hamiltonian overlap operator for each integer $k \geq 0$, with an exactly conserved but nonquantized chiral charge that becomes quantized as $k \to \infty$. Our construction provides physical insight into how Fujikawa's higher-order Ginsparg-Wilson relation improves chiral symmetry while reproducing the anomaly, highlighting the trade-offs inherent in any Hamiltonian lattice realization of an anomalous chiral symmetry. This class of Hamiltonian lattice regularizations, with their tunable chiral symmetry properties, offers potential advantages for quantum and tensor-network simulations.

Singh, Hersh [Fermilab]↗
DOE OSTI2024

Graph characterization of higher-order structure in atmospheric chemical reaction mechanisms

Atmospheric chemical reactions play an important role in air quality and climate change. While the structure and dynamics of individual chemical reactions are fairly well understood, the emergent properties of the entire atmospheric chemical system, which can involve many different species that participate in many different reactions, are not well described. In this work, we leverage graph-theoretic techniques to characterize patterns of interaction (“motifs”) in three different representations of gas-phase atmospheric chemistry, termed “chemical mechanisms.” These widely used mechanisms, the master chemical mechanism, the GEOS-Chem mechanism, and the Super-Fast mechanism, vary dramatically in scale and application, but they all generally aim to simulate the abundance and variability of chemical species in the atmosphere. This motif analysis quantifies the fundamental patterns of interaction within the mechanisms, which are directly related to their construction. For example, the gas-phase chemistry in the very small Super-Fast mechanism is entirely composed of bimolecular reactions, and its motif distribution matches that of an individual bimolecular reaction well. The larger and more complex mechanisms show emergent motif distributions that differ strongly from any specific reaction type, consistent with their complexity. The proposed motif analysis demonstrates that while these mechanisms all have a similar design goal, their higher-order structure of interactions differs strongly and thus provides a novel set of tools for exploring differences across chemical mechanisms.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS1976

A higher order panel method for general analysis and design applications in subsonic flow

A higher-order panel method is described for numerical solution of boundary-value problems relating to steady inviscid irrotational incompressible subsonic fluid flow in a domain. Both Neumann and Dirichlet boundary conditions are treated; two types of auxiliary conditions are used to remove the degrees of freedom that arise from specifying only the derivative of the perturbation velocity potential. Four general network types and two expansions of the induced potential kernel are employed in the numerical solution. Some results are presented which illustrate the modeling options and numerical characteristics of the method.

Johnson, F. T.↗
DOE OSTI2025

Analyzing LF/VLF Lightning Waveforms to Estimate D-region Electron Density Profiles

Lightning waveforms in the low frequency (LF; 30-300 kHz) and the very low frequency (VLF; 3-30 kHz) bands can be exploited to produce data-driven ionospheric D-region electron density profile (EDP) estimates with significantly higher spatial and temporal coverage than previously available. The lightning waveforms used in this paper are signals detected in the LF/VLF of negative cloud-to-ground lightning by the Earth Networks Total Lightning Detection Network. Each waveform contains a ground wave and a time-delayed ionospheric reflection. The time delay between the ground wave and ionospheric reflection has previously been used to estimate a single specular reflection altitude, where LF/VLF emissions are reflected by the ionosphere. Here, we expand upon previous methods to include filtering and spectral analysis, and account for oblique propagation to produce higher-order estimates for reflection altitudes and corresponding electron densities. Once estimated, reflection altitudes and corresponding electron densities can be used to derive parameters β and h’, which define an EDP for the D-region. In this study, the lightning waveform (LW) analysis is demonstrated using a single representative 24-hour dataset over the Southeast United States, and then extended to a total of 10 separate datasets with varying locations and ionospheric conditions. The LW-derived D-region EDPs are in agreement with predictions made by the Faraday International Reference Ionosphere model, and the LW EDPs β and h’ values are consistent with previous LF/VLF-derived estimates.

D-region ionosphere↗
DOE OSTI2025

Neural chaos: A spectral stochastic neural operator

Building surrogate models for operators with uncertainty quantification capabilities is essential for many engineering applications where randomness–such as variability in material properties, boundary conditions, and initial conditions–is unavoidable. Polynomial Chaos Expansion (PCE) is widely recognized as a go-to method for constructing stochastic surrogates in both intrusive and non-intrusive ways, and it has recently been used in the context of operator learning. However, its application becomes challenging for complex or high-dimensional processes, as achieving accuracy requires higher-order polynomials, which can increase computational demand and/or the risk of overfitting. Furthermore, PCE requires specialized treatments to manage random variables that are not independent, and these treatments may be problem-dependent or may fail with increasing complexity. Here, in this work, we adopt the same formalism as the spectral expansion used in PCE; however, we replace the classical polynomial basis functions with neural network (NN) basis functions to leverage their expressivity. To achieve this, we propose an algorithm that identifies NN-parameterized basis functions in a purely data-driven manner, without any prior assumptions about the joint distribution of the random variables involved, whether independent or dependent, or about their marginal distributions. The proposed algorithm identifies each NN-parameterized basis function sequentially, ensuring they are orthogonal with respect to the data distribution. The basis functions are constructed directly on the joint stochastic variables without requiring a tensor product structure or assuming independence of the random variables. This approach may offer greater flexibility for complex stochastic models, while simplifying implementation compared to the tensor product structures typically used in PCE to handle random vectors. This is particularly advantageous given the current state of open-source packages, where building and training neural networks can be done with just a few lines of code and extensive community support. We demonstrate the effectiveness of the proposed scheme through several numerical examples of varying complexity and provide comparisons with classical PCE.

Polynomial chaos expansion↗
DOE OSTI2025

Evaluating Probabilistic Deep Learning Methods for Uncertainty Quantification of Precipitation Bias Correction

Climate models often exhibit biases in their precipitation predictions, particularly underestimating high-intensity events and overestimating low precipitation. Deep learning approaches offer promising solutions, but their epistemic uncertainty associated with a deep learning–based bias correction method has not previously been quantified for reliable downstream climate impact studies. While methods for capturing the epistemic uncertainty in deep learning frameworks exist, there is currently no consensus on the best method. In this work, we compare three uncertainty quantification (UQ) methods—Deep Ensembles (DEns), Monte Carlo Dropout (MCD), and Flipout—by assessing the reliability of their uncertainty estimates using standard measures such as sharpness and calibration. These UQ methods are applied to an existing deep learning precipitation bias correction model known as UFNet: a coupled U-Net and fully connected neural network. The methods utilized to assess the models’ uncertainties are 1) calibration, which ensures that the expected probabilities of the model align with reality and 2) sharpness, which is a measure of the precision of the model’s probabilistic predictions. Of the three UQ methods evaluated, the DEns and MCD methods demonstrated the best-calibrated performance (expected calibration error of 0.36 and 0.35, respectively), compared to Flipout (0.58). In contrast, Flipout had the sharpest predictions and the highest metric performance in bias correcting precipitation—especially for higher-order moments such as kurtosis with a spatial correlation of 72% compared to 32% and 55% spatial correlation for DEns and MCD, respectively. Of the three UQ methods, MCD was found to be the most suitable method for UQ purposes based on its calibration, sharpness, and computational requirements.

Bayesian methods↗
DOE OSTI2026

Learning interpretable surface elasticity properties from bulk properties via neural network equation learners

Surface elasticity is central to understanding the mechanics and stability of surfaces and interfaces. It is characterized by quantities such as surface tension, residual surface stress, and surface stiffness. However their analytical expressions are typically difficult to derive from atomistic data, and depend strongly on modeling choices. This work presents a neural network-based equation learner which combines customized activation functions and connection-based pruning to discover parsimonious, closed-form equations for surface elasticity from atomistic simulations. Applying the method to seven face-centered cubic (FCC) metals, our equation learner uncovers interpretable equations that describe both low-Miller index and high-Miller index surface properties, capturing long-tail property distributions accurately. The discovered expressions are decoupled into two components: a universal, geometry-driven orientation function, and material-specific baseline coefficients. We find that lower-order properties such as surface tension are fundamentally geometry dependent, while higher-order properties such as surface stress and elasticity show more complex geometry and material dependence. We also relate material dependent coefficients to bulk properties, forming a clear map from bulk material properties to surface elasticity. Overall, this approach demonstrates that interpretable neurosymbolic machine learning can bridge the gap between atomistic simulations and physical laws, enabling the discovery of generalizable structure–property relationships for materials science phenomena such as surface elasticity.

Equation learning↗
DOE OSTI2025

Quantum real-time evolution using tensor renormalization group methods

We introduce an approach for approximate real-time evolution of quantum systems using tensor renormalization group (TRG) methods originally developed for imaginary time. We use higher-order TRG to generate a coarse-grained time evolution operator for a 1+1⁢D transverse Ising model with a longitudinal field. We show that the standard tensor norm used for the singular value decomposition-based truncation is degenerate and propose an alternate method to discriminate. We show that it is effective and efficient in evolving Gaussian wave packets for one and two particles in the disordered phase, while ordered phase behavior is more challenging to capture. We compare our algorithm with local simulators for universal quantum computers and discuss possible benchmarking in the near future.

lattice gauge theory↗
DOE OSTI2025

SymProp: Scaling Sparse Symmetric Tucker Decomposition via Symmetry Propagation

Sparse symmetric tensors are an important class of tensors, and their decompositions serve as powerful tools for revealing low-rank structures. This paper introduces SymProp, a novel approach for scaling sparse symmetric Tucker decomposition by propagating symmetry through intermediate computations. SymProp optimizes two key computational kernels: Sparse Symmetric Tensor Times Same Matrix chain (S3 TTMc) for Higher-Order Orthogonal Iteration (HOOI) and Sparse Symmetric Tensor Times Same Matrix chain Times Core (S3 TTMcTC) for Higher-Order QR Iteration (HOQRI). Our method employs a metaprogramming-based index iteration approach to efficiently handle the upper triangular parts of intermediate dense symmetric tensors. SymProp achieves up to 50.9× speedup over SPLATT and up to 360.8× over Compressed Sparse Symmetric (CSS) format on the S3 TTMc operation. Moreover, our S3 TTMc and S3 TTMcTC implementations support tensor orders four levels higher than state-of-the-art methods. Our HOQRI demonstrates superior scalability and up to a 33.6× speedup over optimized HOOI. By enabling more scalable Tucker decompositions for higher orders, decomposition ranks, and dimension sizes, SymProp opens new possibilities for analyzing complex hypergraph structures in fields such as network science, data mining, and machine learning.

Li, Zecheng [North Carolina State University]↗