Search NASA⌕ Search

SEARCH · Search NASA

Results for “variational optimization”

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 505 records · Page 28

Doping evolution of the second magnetization peak and magnetic relaxation in ( B a 1 - x K x ) F e 2 A s 2 single crystals

Here, we present a thorough study of doping dependent magnetic hysteresis and relaxation characteristics in single crystals of (Ba 1-x K x ) Fe 2 As 2 (0.18 ≤ x ≤ 1). The critical current density J c reaches maximum in the underdoped sample x = 0.26 and then decreases in the optimally doped and overdoped samples. Meanwhile, the magnetic relaxation rate S rapidly increases and the flux creep activation barrier U 0 sharply decreases in the overdoped sample x = 0.70. These results suggest that vortex pinning is very strong in the underdoped regime, but it is greatly reduced in the optimally doped and overdoped regime. Transmission electron microscope (TEM) measurements reveal the existence of dislocations and inclusions in all three studied samples x = 0.38, 0.46, and 0.65. An investigation of the paramagnetic Meissner effect (PME) suggests that spatial variations in T c become small in the samples x = 0.43 and 0.46, slightly above the optimal doping levels. Our results support that two types of pinning sources dominate the (Ba 1-x K x ) Fe 2 As 2 crystals: (i) strong δl pinning, which results from the fluctuations in the mean free path l and δT c pinning from the spatial variations in T c in the underdoped regime, and (ii) weak δT c pinning in the optimally doped and overdoped regime.

36 MATERIALS SCIENCE↗

Challenges for unsupervised anomaly detection in particle physics

Anomaly detection relies on designing a score to determine whether a particular event is uncharacteristic of a given background distribution. One way to define a score is to use autoencoders, which rely on the ability to reconstruct certain types of data (background) but not others (signals). In this paper, we study some challenges associated with variational autoencoders, such as the dependence on hyperparameters and the metric used, in the context of anomalous signal (top and W) jets in a QCD background. We find that the hyperparameter choices strongly affect the network performance and that the optimal parameters for one signal are non-optimal for another. In exploring the networks, we uncover a connection between the latent space of a variational autoencoder trained using mean-squared-error and the optimal transport distances within the dataset. We then show that optimal transport distances to representative events in the background dataset can be used directly for anomaly detection, with performance comparable to the autoencoders. Whether using autoencoders or optimal transport distances for anomaly detection, we find that the choices that best represent the background are not necessarily best for signal identification. These challenges with unsupervised anomaly detection bolster the case for additional exploration of semi-supervised or alternative approaches.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

SU-E-T-163: Evaluation of Dose Distributions Recalculated with Per-Field Measurement Data Under the Condition of Respiratory Motion During IMRT for Liver Cancer

Purpose: The dose distributions within the real volumes of tumor targets and critical organs during internal target volume-based intensity-modulated radiation therapy (ITV-IMRT) for liver cancer were recalculated by applying the effects of actual respiratory organ motion, and the dosimetric features were analyzed through comparison with gating IMRT (Gate-IMRT) plan results. Methods: The 4DCT data for 10 patients who had been treated with Gate-IMRT for liver cancer were selected to create ITV-IMRT plans. The ITV was created using MIM software, and a moving phantom was used to simulate respiratory motion. The period and range of respiratory motion were recorded in all patients from 4DCT-generated movie data, and the same period and range were applied when operating the dynamic phantom to realize coincident respiratory conditions in each patient. The doses were recalculated with a 3 dose-volume histogram (3DVH) program based on the per-field data measured with a MapCHECK2 2-dimensional diode detector array and compared with the DVHs calculated for the Gate-IMRT plan. Results: Although a sufficient prescription dose covered the PTV during ITV-IMRT delivery, the dose homogeneity in the PTV was inferior to that with the Gate-IMRT plan. We confirmed that there were higher doses to the organs-at-risk (OARs) with ITV-IMRT, as expected when using an enlarged field, but the increased dose to the spinal cord was not significant and the increased doses to the liver and kidney could be considered as minor when the reinforced constraints were applied during IMRT plan optimization. Conclusion: Because Gate-IMRT cannot always be considered an ideal method with which to correct the respiratory motional effect, given the dosimetric variations in the gating system application and the increased treatment time, a prior analysis for optimal IMRT method selection should be performed while considering the patient's respiratory condition and IMRT plan results.

60 APPLIED LIFE SCIENCES↗

Effect of barren plateaus on gradient-free optimization

Barren plateau landscapes correspond to gradients that vanish exponentially in the number of qubits. Such landscapes have been demonstrated for variational quantum algorithms and quantum neural networks with either deep circuits or global cost functions. For obvious reasons, it is expected that gradient-based optimizers will be significantly affected by barren plateaus. However, whether or not gradient-free optimizers are impacted is a topic of debate, with some arguing that gradient-free approaches are unaffected by barren plateaus. Here we show that, indeed, gradient-free optimizers do not solve the barren plateau problem. Our main result proves that cost function differences, which are the basis for making decisions in a gradient-free optimization, are exponentially suppressed in a barren plateau. Hence, without exponential precision, gradient-free optimizers will not make progress in the optimization. We numerically confirm this by training in a barren plateau with several gradient-free optimizers (Nelder-Mead, Powell, and COBYLA algorithms), and show that the numbers of shots required in the optimization grows exponentially with the number of qubits.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A Toolkit to Study Sensitivity of the Geant4 Predictions to the Variations of the Physics Model Parameters

Geant4 is the leading detector simulation toolkit used in high energy physics to design detectors and to optimize calibration and reconstruction software. It employs a set of carefully validated physics models to simulate interactions of particles with matter across a wide range of interaction energies. These models, especially the hadronic ones, rely largely on directly measured cross-sections and phenomenological predictions with physically motivated parameters estimated by theoretical calculation or measurement. Because these models are tuned to cover a very wide range of possible simulation tasks, they may not always be optimized for a given process or a given material. This raises several critical questions, e.g. how sensitive Geant4 predictions are to the variations of the model parameters, or what uncertainties are associated with a particular tune of a Geant4 physics model, or a group of models, or how to consistently derive guidance for Geant4 model development and improvement from a wide range of available experimental data. We have designed and implemented a comprehensive, modular, user-friendly software toolkit to study and address such questions. It allows one to easily modify parameters of one or several Geant4 physics models involved in the simulation, and to perform collective analysis of multiple variants of the resulting physics observables of interest and comparison against a variety of corresponding experimental data. Based on modern event-processing infrastructure software, the toolkit offers a variety of attractive features, e.g. flexible run-time configurable workflow, comprehensive bookkeeping, easy to expand collection of analytical components. Design, implementation technology, and key functionalities of the toolkit are presented and illustrated with results obtained with Geant4 key hadronic models.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Learning protocols for the fast and efficient control of active matter

Exact analytic calculation shows that optimal control protocols for passive molecular systems often involve rapid variations and discontinuities. However, similar analytic baselines are not generally available for active-matter systems, because it is more difficult to treat active systems exactly. Here we use machine learning to derive efficient control protocols for active-matter systems, and find that they are characterized by sharp features similar to those seen in passive systems. We show that it is possible to learn protocols that effect fast and efficient state-to-state transformations in simulation models of active particles by encoding the protocol in the form of a neural network. We use evolutionary methods to identify protocols that take active particles from one steady state to another, as quickly as possible or with as little energy expended as possible. Our results show that protocols identified by a flexible neural-network ansatz, which allows the optimization of multiple control parameters and the emergence of sharp features, are more efficient than protocols derived recently by constrained analytical methods. Our learning scheme is straightforward to use in experiment, suggesting a way of designing protocols for the efficient manipulation of active matter in the laboratory.

74 ATOMIC AND MOLECULAR PHYSICS↗

ORCHIDEE-PEAT (revision 4596), a model for northern peatland CO 2 , water, and energy fluxes on daily to annual scales

Peatlands store substantial amounts of carbon and are vulnerable to climate change. We present a modified version of the Organising Carbon and Hydrology In Dynamic Ecosystems (ORCHIDEE) land surface model for simulating the hydrology, surface energy, and CO 2 fluxes of peatlands on daily to annual timescales. The model includes a separate soil tile in each 0.5° grid cell, defined from a global peatland map and identified with peat-specific soil hydraulic properties. Runoff from non-peat vegetation within a grid cell containing a fraction of peat is routed to this peat soil tile, which maintains shallow water tables. The water table position separates oxic from anoxic decomposition. The model was evaluated against eddy-covariance (EC) observations from 30 northern peatland sites, with the maximum rate of carboxylation ( V cmax ) being optimized at each site. Regarding short-term day-to-day variations, the model performance was good for gross primary production (GPP) ( r 2 = 0.76; Nash–Sutcliffe modeling efficiency, MEF =0.76) and ecosystem respiration (ER, r 2 = 0.78, MEF = 0.75), with lesser accuracy for latent heat fluxes (LE, r 2 2 exchange (NEE, r 2 = 0.38, MEF = 0.26). Seasonal variations in GPP, ER, NEE, and energy fluxes on monthly scales showed moderate to high r 2 values (0.57–0.86). For spatial across-site gradients of annual mean GPP, ER, NEE, and LE, r 2 values of 0.93, 0.89, 0.27, and 0.71 were achieved, respectively. Water table (WT) variation was not well predicted ( r 2 < 0.1), likely due to the uncertain water input to the peat from surrounding areas. However, the poor performance of WT simulation did not greatly affect predictions of ER and NEE. We found a significant relationship between optimized V cmax and latitude (temperature), which better reflects the spatial gradients of annual NEE than using an average V cmax value

58 GEOSCIENCES↗

Predicting variations of the least principal stress with depth: Application to unconventional oil and gas reservoirs using a log-based viscoelastic stress relaxation model

Knowledge of layer-to-layer variations of the least principal stress, S hmin , with depth is essential for optimization of multi-stage hydraulic fracturing in unconventional reservoirs. Utilizing a geomechanical model based on viscoelastic stress relaxation in relatively clay rich rocks, we present a new method for predicting continuous S hmin variations with depth. The method utilizes geophysical log data and S hmin measurements from routine diagnostic fracture injection tests (DFITs) at several depths for calibration. We consider a case study in the Wolfcamp formation in the Midland Basin, where both geophysical logs and values of S hmin from DFITs are available. We compute a continuous stress profile as a function of the well logs that fits all of the DFITs well. We utilized several machine learning technologies, such as bootstrap aggregation (or bagging), to improve the generalization of the model and demonstrate that the excellent fit between predicted and observed stress values is not the result of over-fitting the calibration points. The model is then validated by accurately predicting hold-out stress measurements from four wells within the study area and, without recalibration, accurately predicting stress as a function of depth in an offset pad about 6 miles away.

58 GEOSCIENCES↗

Proximal Galerkin: A Structure-Preserving Finite Element Method for Pointwise Bound Constraints

The proximal Galerkin finite element method is a high-order, low iteration complexity, nonlinear numerical method that preserves the geometric and algebraic structure of pointwise bound constraints in infinite-dimensional function spaces. This paper introduces the proximal Galerkin method and applies it to solve free boundary problems, enforce discrete maximum principles, and develop a scalable, mesh-independent algorithm for optimal design with pointwise bound constraints. This paper also introduces the latent variable proximal point (LVPP) algorithm, from which the proximal Galerkin method derives. When analyzing the classical obstacle problem, we discover that the underlying variational inequality can be replaced by a sequence of second-order partial differential equations (PDEs) that are readily discretized and solved with, e.g., the proximal Galerkin method. Throughout this work, we arrive at several contributions that may be of independent interest. These include (1) a semilinear PDE we refer to as the entropic Poisson equation; (2) an algebraic/geometric connection between high-order positivity-preserving discretizations and certain infinite-dimensional Lie groups; and (3) a gradient-based, bound-preserving algorithm for two-field, density-based topology optimization. The complete proximal Galerkin methodology combines ideas from nonlinear programming, functional analysis, tropical algebra, and differential geometry and can potentially lead to new synergies among these areas as well as within variational and numerical analysis. Open-source implementations of our methods accompany this work to facilitate reproduction and broader adoption.

97 MATHEMATICS AND COMPUTING↗

TU-H-CAMPUS-TeP1-01: Variable-Beam Fractionation for SAbR

Purpose: In current conventionally-fractionated as well as hypofractionated 3D conformal radiotherapy (CRT), the same beam arrangement is employed from fraction to fraction. We challenge this notion and postulate that by varying the beam arrangement between fractions we can achieve greater sparing of organs at risk (OARs) while maintaining PTV coverage. We use an inverse planning strategy using a swarm intelligence-based global optimization algorithm to exploit the additional degree of freedom represented by inter-fractional variation in beam angles. Methods: To evaluate our variable-beam fractionation (VBF) method, a 10-beam ITV-based conformal stereotactic ablative radiotherapy (CRT-SAbR) plan was optimized. In the clinical plan, 54 Gy was delivered to a 41cc lung tumor over 3 fractions. In VBF, each original clinically-assigned beam was multiplied to a bundle of n α-degree-spaced beams, n being number of fractions. Selection of α was a compromise between retaining similar tumor irradiation and separating inline OAR sub-regions. We optimized the beam fluence weights setting an upper limit for beam delivery duration (and implicitly, monitor units) along with clinical organ-based dose-volume constraints. Zero weights were allowed so that the optimization algorithm could remove unnecessary beams. All fractions in final plan had to deliver identical monitor units (MU) while satisfying a soft constraint on having no more than one beam from every n-beam bundle in each fraction. α was 10 degrees and the dose rate was 600 MU/min. Results: The VBF plan achieved significantly superior OAR sparing compared to the clinical internal target volume (ITV)-based plan. Setting maximum beam delivery duration to 13 seconds (well within breath-hold range), Esophagus Dmax, Heart Dmax, Spinal cord Dmax and Lung V13 were improved by 25%, 81%, 0% and 27%, respectively. Conclusion: We investigated a simple approach to inter-fractional VBF planning and demonstrated its potential in reducing dose to OARs. This work was partially supported through research funding from National Institutes of Health (R01CA169102) and Varian Medical Systems, Palo Alto, CA, USA.

60 APPLIED LIFE SCIENCES↗

Optimizing temperature distributions for training neural quantum states using parallel tempering

Parametrized artificial neural networks (ANNs) can be very expressive ansatzes for variational algorithms, reaching state-of-the-art energies on many quantum many-body Hamiltonians. Nevertheless, the training of the ANN can be slow and stymied by the presence of local minima in the parameter landscape. One approach to mitigate this issue is to use parallel tempering methods, and in this work, we focus on the role played by the temperature distribution of the parallel tempering replicas. Using an adaptive method that adjusts the temperatures in order to equate the exchange probability between neighboring replicas, we show that this temperature optimization can significantly increase the success rate of the variational algorithm with negligible computational cost by eliminating bottlenecks in the replicas' random walk. Furthermore, we demonstrate this using two different neural networks, a restricted Boltzmann machine and a feedforward network, which we use to study a toy problem based on a permutation invariant Hamiltonian with a pernicious local minimum and the 𝐽 1 −𝐽 2 model on a rectangular lattice.

Neural network simulations↗

Imaging the Schwarzschild-radius-scale Structure of M87 with the Event Horizon Telescope Using Sparse Modeling

We propose a new imaging technique for radio and optical/infrared interferometry. The proposed technique reconstructs the image from the visibility amplitude and closure phase, which are standard data products of short-millimeter very long baseline interferometers such as the Event Horizon Telescope (EHT) and optical/infrared interferometers, by utilizing two regularization functions: the ℓ {sub 1}-norm and total variation (TV) of the brightness distribution. In the proposed method, optimal regularization parameters, which represent the sparseness and effective spatial resolution of the image, are derived from data themselves using cross-validation (CV). As an application of this technique, we present simulated observations of M87 with the EHT based on four physically motivated models. We confirm that ℓ {sub 1} + TV regularization can achieve an optimal resolution of ∼20%–30% of the diffraction limit λ / D {sub max}, which is the nominal spatial resolution of a radio interferometer. With the proposed technique, the EHT can robustly and reasonably achieve super-resolution sufficient to clearly resolve the black hole shadow. These results make it promising for the EHT to provide an unprecedented view of the event-horizon-scale structure in the vicinity of the supermassive black hole in M87 and also the Galactic center Sgr A*.

79 ASTRONOMY AND ASTROPHYSICS↗

The Finite Horizon Optimal Multi-Modes Switching Problem: The Viscosity Solution Approach

In this paper we show existence and uniqueness of a solution for a system of m variational partial differential inequalities with inter-connected obstacles. This system is the deterministic version of the Verification Theorem of the Markovian optimal m-states switching problem. The switching cost functions are arbitrary. This problem is in relation with the valuation of firms in a financial market.

97 MATHEMATICS AND COMPUTING↗

Optimization of data collection taking radiation damage into account

Software implementing a new method for the optimal choice of data-collection parameters, accounting for the effects of radiation damage, is presented. To take into account the effects of radiation damage, new algorithms for the optimization of data-collection strategies have been implemented in the software package BEST. The intensity variation related to radiation damage is approximated by log-linear functions of resolution and cumulative X-ray dose. Based on an accurate prediction of the basic characteristics of data yet to be collected, BEST establishes objective relationships between the accessible data completeness, resolution and signal-to-noise statistics that can be achieved in an experiment and designs an optimal plan for data collection.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

SU-F-T-345: Quasi-Dead Beams: Clinical Relevance and Implications for Automatic Planning

Purpose: Beam direction selection for fixed-beam IMRT planning is typically a manual process. Severe dose-volume limits on critical structures in the thorax often result in atypical selection of beam directions as compared to other body sites. This work demonstrates the potential consequences as well as clinical relevance. Methods: 21 thoracic cases treated with 5–7 beam directions, 6 cases including non-coplanar arrangements, with fractional doses of 150–411cGy were analyzed. Endpoints included per-beam modulation scaling factor (MSF), variation from equal weighting, and delivery QA passing rate. Results: During analysis of patient-specific delivery QA a sub-standard passing rate was found for a single 5-field plan (90.48% of pixels evaluated passing 3% dose, 3mm DTA). During investigation it was found that a single beam demonstrated a MSF of 34.7 and contributed only 2.7% to the mean dose of the target. In addition, the variation from equal weighting for this beam was 17.3% absolute resulting in another beam with a MSF of 4.6 contributing 41.9% to the mean dose to the target; a variation of 21.9% from equal weighting. The average MSF for the remaining 20 cases was 4.0 (SD 1.8) with an average absolute deviation of 2.8% from equal weighting (SD 3.1%). Conclusion: Optimization in commercial treatment planning systems typically results in relatively equally weighted beams. Extreme variation from this can result in excessively high MSFs (very small segments) and potential decreases in agreement between planned and delivered dose distributions. In addition, the resultant beam may contribute minimal dose to the target (quasi-dead beam); a byproduct being increased treatment time and associated localization uncertainties. Potential ramifications exist for automatic planning algorithms should they allow for user-defined beam directions. Additionally, these quasi-dead beams may be embedded in the libraries for model-based systems potentially resulting in inefficient and less accurate deliveries.

60 APPLIED LIFE SCIENCES↗

Data-driven optimization of dynamic reconfigurable systems of systems.

This report documents the results of a Strategic Partnership (aka University Collaboration) LDRD program between Sandia National Laboratories and the University of Illinois at Urbana-Champagne. The project is titled 'Data-Driven Optimization of Dynamic Reconfigurable Systems of Systems' and was conducted during FY 2009 and FY 2010. The purpose of this study was to determine and implement ways to incorporate real-time data mining and information discovery into existing Systems of Systems (SoS) modeling capabilities. Current SoS modeling is typically conducted in an iterative manner in which replications are carried out in order to quantify variation in the simulation results. The expense of many replications for large simulations, especially when considering the need for optimization, sensitivity analysis, and uncertainty quantification, can be prohibitive. In addition, extracting useful information from the resulting large datasets is a challenging task. This work demonstrates methods of identifying trends and other forms of information in datasets that can be used on a wide range of applications such as quantifying the strength of various inputs on outputs, identifying the sources of variation in the simulation, and potentially steering an optimization process for improved efficiency.

99 GENERAL AND MISCELLANEOUS↗

High‐Efficiency Nonfullerene Organic Solar Cells: Critical Factors that Affect Complex Multi‐Length Scale Morphology and Device Performance

Organic solar cells (OSCs) made of donor/acceptor bulk‐heterojunction active layers have been of widespread interest in converting sunlight to electricity. Characterizing of the complex morphology at multiple length scales of polymer:nonfullerene small molecular acceptor (SMA) systems remains largely unexplored. Through detailed characterizations (hard/soft X‐ray scattering) of the record‐efficiency polymer:SMA system with a close analog, quantitative morphological parameters are related to the device performance parameters and fundamental morphology–performance relationships that explain why additive use and thermal annealing are needed for optimized performance are established. A linear correlation between the average purity variations at small length scale (≈10 nm) and photovoltaic device characteristics across all processing protocols is observed in ≈12%‐efficiency polymer:SMA systems. In addition, molecular interactions as reflected by the estimated Flory–Huggins interaction parameters are used to provide context of the room temperature morphology results. Comparison with results from annealed devices suggests that the two SMA systems compared show upper and lower critical solution temperature behavior, respectively. The in‐depth understanding of the complex multilength scale nonfullerene OSC morphology may guide the device optimization and new materials development and indicates that thermodynamic properties of materials systems should be studied in more detail to aid in designing optimized protocols efficiently.

Ye, Long↗