Search NASA⌕ Search

SEARCH · Search NASA

Results for “hybrid algorithm”

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 235 records · Page 13

Automated Hybrid Variance Reduction on Advanced Architectures in the Shift Monte Carlo Code

Monte Carlo transport methods are the most accurate schemes for solving problems with complex energy and spatial features, but they come with a high computational cost. Although hybrid methods have enabled the use of Monte Carlo transport for a large class of problems, they still require significant computing resources. Modern multicore CPUs with large numbers of compute cores and graphical processing units (GPUs) provide opportunities to optimize the memory and run-time costs of hybrid Monte Carlo methods. This paper documents the development and analysis of three Monte Carlo transport algorithms that support hybrid transport using the consistent adjoint-driven importance sampling (CADIS) and forward-weighted CADIS methods in the Shift Monte Carlo code: history-based transport using static and dynamic threading on multicore CPUs and event-based transport enabling weight window tracking on GPUs. The results are shown for two challenging hybrid problems on the Frontier supercomputer at the Oak Ridge Leadership Computing Facility. The results show that all three methods yield good performance and enable solutions of difficult fixed-source transport problems in less than 2 min on 20 nodes of Frontier. Dynamic threading was observed to give up to 20% better scaling behavior than static threading. Moreover, the AMD Instinct 250X GPU was found to give 9 to 11 times greater throughput per graphics compute die than the best CPU performance. In conclusion, additional opportunities for optimization of hybrid transport on GPUs are discussed.

Denovo↗

ON THE ESTIMATION OF RANDOM UNCERTAINTIES OF STAR FORMATION HISTORIES

The standard technique for measurement of random uncertainties of star formation histories (SFHs) is the bootstrap Monte Carlo, in which the color-magnitude diagram (CMD) is repeatedly resampled. The variation in SFHs measured from the resampled CMDs is assumed to represent the random uncertainty in the SFH measured from the original data. However, this technique systematically and significantly underestimates the uncertainties for times in which the measured star formation rate is low or zero, leading to overly (and incorrectly) high confidence in that measurement. This study proposes an alternative technique, the Markov Chain Monte Carlo (MCMC), which samples the probability distribution of the parameters used in the original solution to directly estimate confidence intervals. While the most commonly used MCMC algorithms are incapable of adequately sampling a probability distribution that can involve thousands of highly correlated dimensions, the Hybrid Monte Carlo algorithm is shown to be extremely effective and efficient for this particular task. Several implementation details, such as the handling of implicit priors created by parameterization of the SFH, are discussed in detail.

79 ASTRONOMY AND ASTROPHYSICS↗

Antiferromagnetic character of the quantum phase transition in the Hubbard model on the honeycomb lattice

In this work we provide a unified, comprehensive treatment of all operators that contribute to the antiferromagnetic, ferromagnetic, and charge-density-wave structure factors and order parameters of the hexagonal Hubbard Model. We use the Hybrid Monte Carlo algorithm to perform a systematic, carefully controlled analysis in the temporal Trotter error and of the thermodynamic limit. We expect our findings to improve the consistency of Monte Carlo determinations of critical exponents. We perform a data collapse analysis and determine the critical exponent β = 0.898 (37) for the semimetal-Mott insulator transition in the hexagonal Hubbard Model. Our methods are applicable to a wide range of lattice theories of strongly correlated electrons.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

An ammonia spectral map of the L1495-B218 filaments in the Taurus molecular cloud. I. Physical properties of filaments and dense cores

We present deep NH{sub 3} observations of the L1495-B218 filaments in the Taurus molecular cloud covering over a 3° angular range using the K-band focal plane array on the 100 m Green Bank Telescope. The L1495-B218 filaments form an interconnected, nearby, large complex extending over 8 pc. We observed NH{sub 3} (1, 1) and (2, 2) with a spectral resolution of 0.038 km s{sup −1} and a spatial resolution of 31″. Most of the ammonia peaks coincide with intensity peaks in dust continuum maps at 350 and 500 μm. We deduced physical properties by fitting a model to the observed spectra. We find gas kinetic temperatures of 8–15 K, velocity dispersions of 0.05–0.25 km s{sup −1}, and NH{sub 3} column densities of 5 × 10{sup 12} to 1 × 10{sup 14} cm{sup −2}. The CSAR algorithm, which is a hybrid of seeded-watershed and binary dendrogram algorithms, identifies a total of 55 NH{sub 3} structures, including 39 leaves and 16 branches. The masses of the NH{sub 3} sources range from 0.05 to 9.5 M{sub ⊙}. The masses of NH{sub 3} leaves are mostly smaller than their corresponding virial mass estimated from their internal and gravitational energies, which suggests that these leaves are gravitationally unbound structures. Nine out of 39 NH{sub 3} leaves are gravitationally bound, and seven out of nine gravitationally bound NH{sub 3} leaves are associated with star formation. We also found that 12 out of 30 gravitationally unbound leaves are pressure confined. Our data suggest that a dense core may form as a pressure-confined structure, evolve to a gravitationally bound core, and undergo collapse to form a protostar.

79 ASTRONOMY AND ASTROPHYSICS↗

A Scalable Hybrid Scheme for Ray-Casting of Unstructured Volume Data

We present an algorithm for parallel volume rendering that is a hybrid between classical object order and image order techniques. The algorithm operates on unstructured grids (and structured ones), and thus can deal with block boundaries interleaving in complex ways. It also deals effectively with cases that are prone to load imbalance, i.e., cases where cell sizes differ dramatically, either because of the nature of the input data, or because of the effects of the camera transformation. The algorithm divides work over resources such that each phase of its processing is bounded in the amount of computation it can perform. We demonstrate its efficacy through a series of studies, varying over camera position, data set size, transfer function, image size, and processor count. At its biggest, our experiments scaled up to 8,192 processors and operated on data sets with more than one billion cells. In total, we find that our hybrid algorithm performs well in all cases. This is because our algorithm naturally adapts its computation based on workload, and can operate like either an object order technique or an image order technique in scenarios where those techniques are efficient.

97 MATHEMATICS AND COMPUTING↗

Accurate and Accelerated Neuromorphic Network Design Leveraging A Bayesian Hyperparameter Pareto Optimization Approach

Neuromorphic systems allow for extremely efficient hardware implementations for neural networks (NNs). In recent years, several algorithms have been presented to train spiking NNs (SNNs) for neuromorphic hardware. However, SNNs often provide lower accuracy than their artificial NNs (ANNs) counterparts or require computationally expensive and slow training/inference methods. To close this gap, designers typically rely on reconfiguring SNNs through adjustments in the neuron/synapse model or training algorithm itself. Nevertheless, these steps incur significant design time, while still lacking the desired improvement in terms of training/inference times (latency). Designing SNNs that can mimic the accuracy of ANNs with reasonable training times is an exigent challenge in neuromorphic computing. In this work, we present an alternative approach that looks at such designs as an optimization problem rather than algorithm or architecture redesign. We develop a versatile multiobjective hyperparameter optimization (HPO) for automatically tuning HPs of two state-of-the-art SNN training algorithms, SLAYER and HYBRID. We emphasize that, to the best of our knowledge, this is the first work trying to improve SNNs’ computational efficiency, accuracy, and training time using an efficient HPO. We demonstrate significant performance improvements for SNNs on several datasets without the need to redesign or invent new training algorithms/architectures. Our approach results in more accurate networks with lower latency and, in turn, higher energy efficiency than previous implementations. In particular, we demonstrate improvement in accuracy and more than 5× reduction in the training/inference time for the SLAYER algorithm on the DVS Gesture dataset. In the case of HYBRID, we demonstrate 30% reduction in timesteps while surpassing the accuracy of the state-of-the-art networks on CIFAR10. Further, our analysis suggests that even a seemingly minor change in HPs could change the accuracy by 5 - 6×.

Parsa, Maryam↗

Automatic Thread-Level Parallelization in the Chombo AMR Library

The increasing on-chip parallelism has some substantial implications for HPC applications. Currently, hybrid programming models (typically MPI+OpenMP) are employed for mapping software to the hardware in order to leverage the hardware?s architectural features. In this paper, we present an approach that automatically introduces thread level parallelism into Chombo, a parallel adaptive mesh refinement framework for finite difference type PDE solvers. In Chombo, core algorithms are specified in the ChomboFortran, a macro language extension to F77 that is part of the Chombo framework. This domain-specific language forms an already used target language for an automatic migration of the large number of existing algorithms into a hybrid MPI+OpenMP implementation. It also provides access to the auto-tuning methodology that enables tuning certain aspects of an algorithm to hardware characteristics. Performance measurements are presented for a few of the most relevant kernels with respect to a specific application benchmark using this technique as well as benchmark results for the entire application. The kernel benchmarks show that, using auto-tuning, up to a factor of 11 in performance was gained with 4 threads with respect to the serial reference implementation.

97 MATHEMATICS AND COMPUTING↗

Multiscale Monte Carlo equilibration: Two-color QCD with two fermion flavors

In this study, we demonstrate the applicability of a recently proposed multiscale thermalization algorithm to two-color quantum chromodynamics (QCD) with two mass-degenerate fermion flavors. The algorithm involves refining an ensemble of gauge configurations that had been generated using a renormalization group (RG) matched coarse action, thereby producing a fine ensemble that is close to the thermalized distribution of a target fine action; the refined ensemble is subsequently rethermalized using conventional algorithms. Although the generalization of this algorithm from pure Yang-Mills theory to QCD with dynamical fermions is straightforward, we find that in the latter case, the method is susceptible to numerical instabilities during the initial stages of rethermalization when using the hybrid Monte Carlo algorithm. We find that these instabilities arise from large fermion forces in the evolution, which are attributed to an accumulation of spurious near-zero modes of the Dirac operator. We propose a simple strategy for curing this problem, and demonstrate that rapid thermalization--as probed by a variety of gluonic and fermionic operators--is possible with the use of this solution. Also, we study the sensitivity of rethermalization rates to the RG matching of the coarse and fine actions, and identify effective matching conditions based on a variety of measured scales.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Feedback-based quantum algorithms for ground state preparation

The ground state properties of quantum many-body systems are a subject of interest across chemistry, materials science, and physics. Thus, algorithms for finding ground states can have broad impacts. Variational quantum algorithms are one class of ground state algorithms that has received significant attention in recent years. These algorithms utilize a hybrid quantum-classical computing framework to prepare ground states on quantum computers. However, this requires solving a classical optimization problem that can become prohibitively expensive in high dimensions. Here, we develop formulations of feedback-based quantum algorithms for ground state preparation that can be used to address this challenge for two broad classes of Hamiltonians: Fermi-Hubbard Hamiltonians, and molecular Hamiltonians represented in second quantization. Feedback-based quantum algorithms are optimization-free; in place of classical optimization, quantum circuit parameters are set according to a deterministic feedback law derived from quantum Lyapunov control principles. This feedback law guarantees a monotonic improvement in solution quality with respect to the depth of the quantum circuit. A variety of numerical illustrations are provided that analyze the convergence and robustness of feedback-based quantum algorithms for these problem classes. Published by the American Physical Society 2024

Larsen, James B. (ORCID:000000020777440X)↗

Modeling L 2,3 -edge X-ray absorption spectroscopy with linear response exact two-component relativistic time-dependent density functional theory

X-ray absorption spectroscopy (XAS) is a powerful tool that can provide physical insights into element-specific chemical processes and reactivities. Although relativistic time-dependent density functional theory (TDDFT) has been previously applied to model the L-edge region in XAS, there has not been a more comprehensive study of the choices of basis sets and density functional kernels available for variational relativistic excited state methods. In this work, we introduce the implementation of the generalized preconditioned locally harmonic residual algorithm to solve the complex-valued relativistic TDDFT for modeling the L-edge X-ray absorption spectra. Further, we investigate the L 2,3 -edge spectra of a series of molecular complexes using relativistic linear response TDDFT with a hybrid iterative diagonalization algorithm. A systematic error analysis was carried out with a focus on the energetics, intensities, and magnitude of L 2 –L 3 splitting compared to experiments. Additionally, the results from relativistic TDDFT calculations are compared to those computed using other theoretical methods, and the multideterminantal effects on the L-edge XAS were investigated.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Characterization of a commercial hybrid iterative and model-based reconstruction algorithm in radiation oncology

Purpose: Iterative reconstruction (IR) reduces noise, thereby allowing dose reduction in computed tomography (CT) while maintaining comparable image quality to filtered back-projection (FBP). This study sought to characterize image quality metrics, delineation, dosimetric assessment, and other aspects necessary to integrate IR into treatment planning. Methods: CT images (Brilliance Big Bore v3.6, Philips Healthcare) were acquired of several phantoms using 120 kVp and 25–800 mAs. IR was applied at levels corresponding to noise reduction of 0.89–0.55 with respect to FBP. Noise power spectrum (NPS) analysis was used to characterize noise magnitude and texture. CT to electron density (CT-ED) curves were generated over all IR levels. Uniformity as well as spatial and low contrast resolution were quantified using a CATPHAN phantom. Task specific modulation transfer functions (MTF{sub task}) were developed to characterize spatial frequency across objects of varied contrast. A prospective dose reduction study was conducted for 14 patients undergoing interfraction CT scans for high-dose rate brachytherapy. Three physicians performed image quality assessment using a six-point grading scale between the normal-dose FBP (reference), low-dose FBP, and low-dose IR scans for the following metrics: image noise, detectability of the vaginal cuff/bladder interface, spatial resolution, texture, segmentation confidence, and overall image quality. Contouring differences between FBP and IR were quantified for the bladder and rectum via overlap indices (OI) and Dice similarity coefficients (DSC). Line profile and region of interest analyses quantified noise and boundary changes. For two subjects, the impact of IR on external beam dose calculation was assessed via gamma analysis and changes in digitally reconstructed radiographs (DRRs) were quantified. Results: NPS showed large reduction in noise magnitude (50%), and a slight spatial frequency shift (∼0.1 mm{sup −1}) with application of IR at L6. No appreciable changes were observed for CT-ED curves between FBP and IR levels [maximum difference ∼13 HU for bone (∼1% difference)]. For uniformity, differences were ∼1 HU between FBP and IR. Spatial resolution was well conserved; the largest MTF{sub task} decrease between FBP and IR levels was 0.08 A.U. No notable changes in low-contrast detectability were observed and CNR increased substantially with IR. For the patient study, qualitative image grading showed low-dose IR was equivalent to or slightly worse than normal dose FBP, and is superior to low-dose FBP (p < 0.001 for noise), although these did not translate to differences in CT number, contouring ability, or dose calculation. The largest CT number discrepancy from FBP occurred at a bone/tissue interface using the most aggressive IR level [−1.2 ± 4.9 HU (range: −17.6–12.5 HU)]. No clinically significant contour differences were found between IR and FBP, with OIs and DSCs ranging from 0.85 to 0.95. Negligible changes in dose calculation were observed. DRRs preserved anatomical detail with <2% difference in intensity from FBP combined with aggressive IRL6. Conclusions: These results support integrating IR into treatment planning. While slight degradation in edges and shift in texture were observed in phantom, patient results show qualitative image grading, contouring ability, and dosimetric parameters were not adversely affected.

60 APPLIED LIFE SCIENCES↗

Optimal Power Dispatch of DGs in Radial and Mesh AC Grids: A Hybrid Solution Methodology between the Salps Swarm Algorithm and Successive Approximation Power Flow Method

In this paper, we address the problem of the optimal power dispatch of Distributed Generators (DGs) in Alternating Current (AC) networks, better known as the Optimal Power Flow (OPF) problem. We used, as the objective function, the minimization of power losses (P loss ) associated with energy transport, which are subject to the set of constraints that compose AC networks in an environment of distributed generation. To validate the effectiveness of the proposed methodology in solving the OPF problem in any network topology, we employed one 10-node mesh test system and three radial text systems: 10, 33, and 69 nodes. In each test system, DGs were allowed to inject 20%, 40%, and 60% of the power supplied by the slack generator in the base case. To solve the OPF problem, we used a master–slave methodology that integrates the optimization method Salps Swarm Algorithm (SSA) and the load flow technique based on the Successive Approximation (SA) method. Moreover, for comparison purposes, we employed some of the algorithms reported in the specialized literature to solve the OPF problem (the continuous genetic algorithm, the particle swarm optimization algorithm, the black hole algorithm, the antlion optimization algorithm, and the Multi-Verse Optimizer algorithm), which were selected because of their excellent results in solving such problems. The results obtained by the proposed solution methodology demonstrate its superiority and convergence capacity in terms of minimization of P loss in both radial and mesh systems. It provided the best reduction in minimum P loss in short processing times and showed excellent repeatability in each test system and scenario under analysis.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Batched matrix computations on hardware accelerators based on GPUs

Scientific applications require solvers that work on many small size problems that are independent from each other. At the same time, the high-end hardware evolves rapidly and becomes ever more throughput-oriented and thus there is an increasing need for an effective approach to develop energy-efficient, high-performance codes for these small matrix problems that we call batched factorizations. The many applications that need this functionality could especially benefit from the use of GPUs, which currently are four to five times more energy efficient than multicore CPUs on important scientific workloads. This study, consequently, describes the development of the most common, one-sided factorizations, Cholesky, LU, and QR, for a set of small dense matrices. The algorithms we present together with their implementations are, by design, inherently parallel. In particular, our approach is based on representing the process as a sequence of batched BLAS routines that are executed entirely on a GPU. Importantly, this is unlike the LAPACK and the hybrid MAGMA factorization algorithms that work under drastically different assumptions of hardware design and efficiency of execution of the various computational kernels involved in the implementation. Thus, our approach is more efficient than what works for a combination of multicore CPUs and GPUs for the problems sizes of interest of the application use cases. The paradigm where upon a single chip (a GPU or a CPU) factorizes a single problem at a time is not at all efficient in our applications’ context. We illustrate all of these claims through a detailed performance analysis. With the help of profiling and tracing tools, we guide our development of batched factorizations to achieve up to two-fold speedup and three-fold better energy efficiency as compared against our highly optimized batched CPU implementations based on MKL library. Finally, the tested system featured two sockets of Intel Sandy Bridge CPUs and we compared with a batched LU factorizations featured in the CUBLAS library for GPUs, we achieve as high as 2.5× speedup on the NVIDIA K40 GPU.

97 MATHEMATICS AND COMPUTING↗

Simultaneous prediction of structural properties in epitaxially–grown GaN with quantum and conventional multi–output learning algorithms

Hundreds of GaN thin film crystal plasma–assisted molecular beam epitaxy synthesis experiment records spanning two decades were organized into a dataset correlating the growth experiment design parameters with discrete, binary determinations of crystallinity and surface morphology. Conventional data science techniques as well as both quantum and classical multi–output supervised machine learning algorithms were implemented to investigate the relationships between the operating parameter data and the structural figures of merit. Correlation coefficients, decision tree nodes, p–values, and SHAP values all support substrate temperature and gallium effusion cell conditions as being statistically significant for simultaneously influencing GaN crystallinity and surface morphology. Here, a conventional deep neural network learned best from the data, followed by a quantum–classical hybrid gradient boosting algorithm. When combined with calculations of uncertainty intervals based on VennAbers predictors, machine learning predictions of both structural properties show good agreement with results reported in published experimental literature.

36 MATERIALS SCIENCE↗

Evolutionary computation algorithms to coordinating order acceptance and batch delivery for an integrated supply chain scheduling

This paper proposes a novel approach of coordinating decisions in an integrated supply chain (ISC): coordinating order acceptance (OA) and batch delivery (BD) due to round trip transportation (RTT) and using third-party logistics (3PL) vehicles. The paper aims at trading off among accepted orders revenue, delivery costs as well as any penalties incurred in the ISC to maximize the total benefit. A novel mixed-integer programming is proposed for the problem. In addition, the paper provides a heuristic to form batches and develops a hybrid evolutionary computation algorithms based on particle swarm optimization (PSO) and genetic algorithm (GA) to solve the problem. An information sharing mechanism is improved and applied. To explore and to locate the proposed PSO in a better neighborhood, a local search is proposed. Taguchi experimental design is utilized to set the appropriate values of the algorithms’ parameters and random instances are generated to evaluate the performance of the algorithms. The paper investigates the profitability sensitivity of the problem to parameters and analyzes the effect of the changes in the parameters on the performance of our proposed algorithms. The attained results show the appropriate performance of our algorithms.

97 MATHEMATICS AND COMPUTING↗

Hybrid feature selection approach for power transformer fault diagnosis based on support vector machine and genetic algorithm

To further improve fault diagnosis accuracy, a new hybrid feature selection approach combined with a genetic algorithm (GA) and support vector machine (SVM) is presented in this study. Adaptive synthetic technique and arctangent transformation method are adopted to improve the statistical property of the training set (IEC TC10 dataset). Five filter methods based on different evaluation metrics are employed to rank 48 input features derived from dissolved gas analysis (DGA). Then, feature combination methods are applied to aggregate feature ranks and form a lower-dimension candidate feature subset. The GA-SVM model is implemented to optimise parameters and select optimal feature subsets. 5-fold cross-validation accuracy of the GA-SVM is used to evaluate fault diagnosis capability of feature subsets and finally, a novel subset is determined as the optimal feature subset. Accuracy comparison manifests the superiority of the optimal feature subsets over that of conventional approaches. Besides, generalisation and robustness of the optimal subset are validated by testing DGA samples from the local power utility. Results indicate that the optimal feature subset obtained by the proposed method can significantly improve the accuracies of power transformer fault diagnosis.

Kari, Tusongjiang↗