Search NASA⌕ Search

SEARCH · Search NASA

Results for “feature selection”

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 829 records · Page 46

MONK10: A Monte Carlo Code for Criticality Analysis - 14574

The MONK{sup R} code, if we include its predecessor GEM, has now been in use for 50 years. It is a Monte Carlo code for nuclear criticality and reactor physics analyses which is used extensively in the UK nuclear industry and worldwide. In 2014 a major new version of the code, MONK10A, was released introducing many new features; these are described with example applications presented where appropriate. Experience with MONK9A models indicates that MONK10A has increased run-time performance for all cases and on average runs 2.4 times faster. New geometry options include the direct import of CAD generated models using the new IGES body, POLY body, TETMESH Hole and TETMESH Zone features. Example calculations for a selection of CAD import options are presented and compared with the intrinsic MONK geometry representation, with relative merits discussed. New Features are included to assist with the modeling of distributed parameters; such features include Unified Tally (UT), Action Tallies (AT) in UT and Shannon entropy. The latter is useful in assessing the convergence of distributed parameters, such as scalar neutron flux. As such it is an important tool in determining the appropriate point to stop settling and begin scoring stages in a Monte Carlo simulation. The utility of UT, AT-in-UT and Shannon entropy is demonstrated using calculations employing the new CAD import geometry options. Runtime Doppler broadening is included to allow accurate representation of temperature distributions within models. Example calculations which further demonstrate UT and AT-in-UT are presented for different bodies at different temperatures to evaluate the effects of temperature differences on k-effective and reaction rates. The MONK10A code was issued with a new set of continuous energy and group nuclear data libraries. These include JEFF3.1, JEFF3.1.1, JEFF 3.1.2, ENDF/B-VII.0 and CENDL 3.1 libraries. The inclusion of a range of libraries allows the user to evaluate the impact of different international nuclear data evaluations on their results and an example is provided to illustrate the differences that can arise from the use of different nuclear data evaluations. The code was also issued with a new version of the VisualWorkshop IDE to allow easy construction and editing of input decks, running of calculations and display of geometry and results. Examples are provided of the geometry and results display and showing how results can be superimposed upon the geometry. (authors)

42 ENGINEERING↗

Prognostic Value and Reproducibility of Pretreatment CT Texture Features in Stage III Non-Small Cell Lung Cancer

Purpose: To determine whether pretreatment CT texture features can improve patient risk stratification beyond conventional prognostic factors (CPFs) in stage III non-small cell lung cancer (NSCLC). Methods and Materials: We retrospectively reviewed 91 cases with stage III NSCLC treated with definitive chemoradiation therapy. All patients underwent pretreatment diagnostic contrast enhanced computed tomography (CE-CT) followed by 4-dimensional CT (4D-CT) for treatment simulation. We used the average-CT and expiratory (T50-CT) images from the 4D-CT along with the CE-CT for texture extraction. Histogram, gradient, co-occurrence, gray tone difference, and filtration-based techniques were used for texture feature extraction. Penalized Cox regression implementing cross-validation was used for covariate selection and modeling. Models incorporating texture features from the 33 image types and CPFs were compared to those with models incorporating CPFs alone for overall survival (OS), local-regional control (LRC), and freedom from distant metastases (FFDM). Predictive Kaplan-Meier curves were generated using leave-one-out cross-validation. Patients were stratified based on whether their predicted outcome was above or below the median. Reproducibility of texture features was evaluated using test-retest scans from independent patients and quantified using concordance correlation coefficients (CCC). We compared models incorporating the reproducibility seen on test-retest scans to our original models and determined the classification reproducibility. Results: Models incorporating both texture features and CPFs demonstrated a significant improvement in risk stratification compared to models using CPFs alone for OS (P=.046), LRC (P=.01), and FFDM (P=.005). The average CCCs were 0.89, 0.91, and 0.67 for texture features extracted from the average-CT, T50-CT, and CE-CT, respectively. Incorporating reproducibility within our models yielded 80.4% (±3.7% SD), 78.3% (±4.0% SD), and 78.8% (±3.9% SD) classification reproducibility in terms of OS, LRC, and FFDM, respectively. Conclusions: Pretreatment tumor texture may provide prognostic information beyond that obtained from CPFs. Models incorporating feature reproducibility achieved classification rates of ∼80%. External validation would be required to establish texture as a prognostic factor.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Feature issue introduction: light, energy and the environment, 2017

The editors introduce the feature issue on 'Energy, Light and the Environment (LEE) 2017', which is based on the topics presented at a congress of the same name held in Boulder, CO, US, from November 6 to November 9. This feature issue presents 13 papers selected from the voluntary submissions by attendees who presented at the progress and have extended their work into complete research articles. Furthermore, the feature issue highlights contributions from authors who presented their research at this congress.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Iterative self-organizing SCEne-LEvel sampling (ISOSCELES) for large-scale building extraction

Convolutional neural networks (CNN) provide state-of-the-art performance in many computer vision tasks, including those related to remote-sensing image analysis. Successfully training a CNN to generalize well to unseen data, however, requires training on samples that represent the full distribution of variation of both the target classes and their surrounding contexts. With remote sensing data, acquiring a sufficiently representative training set is a challenge due to both the inherent multi-modal variability of satellite or aerial imagery and the general high cost of labeling data. To address this challenge, we have developed ISOSCELES, an Iterative Self-Organizing SCEne LEvel Sampling method for hierarchical sampling of large image sets. Using affinity propagation, ISOSCELES automates the selection of highly representative training images. Compared to random sampling or using available reference data, the distribution of the training is principally data driven, reducing the chance of oversampling uninformative areas or undersampling informative ones. In comparison to manual sample selection by an analyst, ISOSCELES exploits descriptive features, spectral and/or textural, and eliminates human bias in sample selection. Using a hierarchical sampling approach, ISOSCELES can obtain a training set that reflects both between-scene variability, such as in viewing angle and time of day, and within-scene variability at the level of individual training samples. We verify the method by demonstrating its superiority to stratified random sampling in the challenging task of adapting a pre-trained model to a new image and spatial domain for country-scale building extraction. Using a pair of hand-labeled training sets comprising 1,987 sample image chips, a total of 496,000,000 individually labeled pixels, we show, across three distinct model architectures, an increase in accuracy, as measured by F1-score, of 2.2–4.2%.

42 ENGINEERING↗

Image Discrimination Models With Stochastic Channel Selection

Many models of human image processing feature a large fixed number of channels representing cortical units varying in spatial position (visual field direction and eccentricity) and spatial frequency (radial frequency and orientation). The values of these parameters are usually sampled at fixed values selected to ensure adequate overlap considering the bandwidth and/or spread parameters, which are usually fixed. Even high levels of overlap does not always ensure that the performance of the model will vary smoothly with image translation or scale changes. Physiological measurements of bandwidth and/or spread parameters result in a broad distribution of estimated parameter values and the prediction of some psychophysical results are facilitated by the assumption that these parameters also take on a range of values. Selecting a sample of channels from a continuum of channels rather than using a fixed set can make model performance vary smoothly with changes in image position, scale, and orientation. It also facilitates the addition of spatial inhomogeneity, nonlinear feature channels, and focus of attention to channel models.

Ahumada, Albert J., Jr.↗

Elemental Analysis of Beryllium Samples Using a Microzond-EGP-10 Unit

Results concerning the structural and elemental analysis of beryllium samples obtained via different technologies using a Microzond-EGP-10 unit with the help of the PIXE and RBS methods are presented. As a result, the overall chemical composition and the nature of inclusions were determined. The mapping method made it possible to reveal the structural features of beryllium samples: to select the grains of the main substance having different size and chemical composition, to visualize the interfaces between the regions of different composition, and to describe the features of the distribution of impurities in the samples.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

SU-F-R-22: Malignancy Classification for Small Pulmonary Nodules with Radiomics and Logistic Regression

Purpose: We conducted a retrospective study of Radiomics research for classifying malignancy of small pulmonary nodules. A machine learning algorithm of logistic regression and open research platform of Radiomics, IBEX (Imaging Biomarker Explorer), were used to evaluate the classification accuracy. Methods: The training set included 100 CT image series from cancer patients with small pulmonary nodules where the average diameter is 1.10 cm. These patients registered at Chang Gung Memorial Hospital and received a CT-guided operation of lung cancer lobectomy. The specimens were classified by experienced pathologists with a B (benign) or M (malignant). CT images with slice thickness of 0.625 mm were acquired from a GE BrightSpeed 16 scanner. The study was formally approved by our institutional internal review board. Nodules were delineated and 374 feature parameters were extracted from IBEX. We first used the t-test and p-value criteria to study which feature can differentiate between group B and M. Then we implemented a logistic regression algorithm to perform nodule malignancy classification. 10-fold cross-validation and the receiver operating characteristic curve (ROC) were used to evaluate the classification accuracy. Finally hierarchical clustering analysis, Spearman rank correlation coefficient, and clustering heat map were used to further study correlation characteristics among different features. Results: 238 features were found differentiable between group B and M based on whether their statistical p-values were less than 0.05. A forward search algorithm was used to select an optimal combination of features for the best classification and 9 features were identified. Our study found the best accuracy of classifying malignancy was 0.79±0.01 with the 10-fold cross-validation. The area under the ROC curve was 0.81±0.02. Conclusion: Benign nodules may be treated as a malignant tumor in low-dose CT and patients may undergo unnecessary surgeries or treatments. Our study may help radiologists to differentiate nodule malignancy for low-dose CT.

60 APPLIED LIFE SCIENCES↗

Using automated machine learning for the upscaling of gross primary productivity

Estimating gross primary productivity (GPP) over space and time is fundamental for understanding the response of the terrestrial biosphere to climate change. Eddy covariance flux towers provide in situ estimates of GPP at the ecosystem scale, but their sparse geographical distribution limits larger-scale inference. Machine learning (ML) techniques have been used to address this problem by extrapolating local GPP measurements over space using satellite remote sensing data. However, the accuracy of the regression model can be affected by uncertainties introduced by model selection, parameterization, and choice of explanatory features, among others. Recent advances in automated ML (AutoML) provide a novel automated way to select and synthesize different ML models. In this work, we explore the potential of AutoML by training three major AutoML frameworks on eddy covariance measurements of GPP at 243 globally distributed sites. We compared their ability to predict GPP and its spatial and temporal variability based on different sets of remote sensing explanatory variables. Explanatory variables from only Moderate Resolution Imaging Spectroradiometer (MODIS) surface reflectance data and photosynthetically active radiation explained over 70 % of the monthly variability in GPP, while satellite-derived proxies for canopy structure, photosynthetic activity, environmental stressors, and meteorological variables from reanalysis (ERA5-Land) further improved the frameworks' predictive ability. We found that the AutoML framework Auto-sklearn consistently outperformed other AutoML frameworks as well as a classical random forest regressor in predicting GPP but with small performance differences, reaching an r 2 of up to 0.75. We deployed the best-performing framework to generate global wall-to-wall maps highlighting GPP patterns in good agreement with satellite-derived reference data. This research benchmarks the application of AutoML in GPP estimation and assesses its potential and limitations in quantifying global photosynthetic activity.

54 ENVIRONMENTAL SCIENCES↗

The crystal structures of PKG Iβ (92-227) with cGMP and cAMP reveal the molecular details of cyclic-nucleotide binding

Cyclic GMP is a crucial second messenger that translates extracellular signals into a variety of cellular responses. As a central mediator of the Nitric Oxide-cGMP signalling cascade, which regulates vascular tone, platelet aggregation, nociception and hipocampal/cerebellar learning, Cyclic GMP-dependent protein kinases (PKGs) represents an important drug target for treating hypertensive diseases and erectile dysfunction. The fidelity of the NO-cGMP signaling pathway is largely dependent on PKG’s ability to selectively bind cGMP over cAMP. Although both cGMP and cAMP bind and activate PKG, cGMP preferentially activates PKG 60-100 fold better than cAMP; yet, little is known about the molecular features required for the cGMP selectivity of PKG. We have investigated the mechanism of cyclic nucleotide binding to PKG by determining crystal structures of the amino-terminal cyclic nucleotide-binding domain (CNBD-A) of human PKG I bound to either cGMP or cAMP. We also determined the structure of CNBD-A in the absence of bound nucleotide.

60 APPLIED LIFE SCIENCES↗

Enhancing and Extinguishing the Different Emission Features of 2D (EA(1-)(x)FA(x))(4)Pb3Br10 Perovskite Films

2D hybrid perovskites are attractive for optoelectronic devices. In thin films, the color of optical emission and the texture of crystalline domains are often difficult to control. Here, a method for extinguishing or enhancing different emission features is demonstrated for the family of 2D Ruddlesden-Popper perovskites (EA(1-)(x)FA(x))(4)Pb3Br10 (EA = ethylammonium, FA = formamidinium). When grown from aqueous hydrobromic acid, crystals of (EA(1-)(x)FA(x))(4)Pb3Br10 retain all the emission features of their parent compound, (EA)(4)Pb3Br10. Surprisingly, when grown from dimethylformamide (DMF), an emission feature, likely self-trapped exciton (STE), near 2.7 eV is missing. Extinction of this feature is correlated with DMF being incorporated between the 2D Pb-Br sheets, forming (EA(1-)(x)FA(x))(4)Pb3Br10 center dot(DMF)(y). Without FA, films grown from DMF form (EA)(4)Pb3Br10, retain little solvent, and have strong emission near 2.7 eV. Slowing the kinetics of film growth strengthens a different emission feature, likely a different type of STE, which is much broader and present in all compositions. Films of (EA(1-)(x)FA(x))(4)Pb3Br10 center dot(DMF)(y) have large, micron-sized domains and homogeneous orientation of the semiconducting sheets, resulting in low electronic disorder near the absorption edge. The ability to selectively strengthen or extinguish different emission features in films of (EA(1-)(x)FA(x))(4)Pb3Br10 center dot(DMF)(y) reveals a pathway to tune the emission color in these compounds.

2D halide perovskites↗

Structural hierarchy as a key to complex phase selection in Al-Sm

Investigating the unknown structure of the complex cubic phase, previously observed to crystallize from melt-spun amorphous Al–10 at.% Sm alloy, we determine the structure in full site-occupancy detail, highlighting several critical structural features that govern the far-from-equilibrium phase selection pathway. Using an efficient genetic algorithm combining molecular dynamics, density functional theory, and x-ray diffraction, the structure is clearly identified as body-centered cubic Im¯3m (No. 229) with ~140 atoms per cubic unit cell and a lattice parameter of 1.4 nm. The complex structure is further refined to elucidate the detailed site occupancy, revealing full Sm occupancy on 6b sites and split Sm/Al occupancy on 16f sites. Based on the refined site occupancy associated with the experimentally observed phase, we term this phase ε–Al 60 Sm 11 (bcc), corresponding to the limiting situation when all 16f sites are occupied by Sm. However, it should be recognized that the range of solubility enabled by split occupancy at Sm sites is an important feature in phase selection under experimental conditions, permitting an avenue for transition with little or no chemical partitioning. Our analysis shows that the ε–Al 60 Sm 11 (bcc) exhibits a “3-6-6-1” first-shell packing around Sm centers on 16f sites, the same dominant motif exhibited by the undercooled liquid. Here, the coincident motif supports the notion that liquid/glass ordering at high undercooling may give rise to topological invariants between the noncrystalline and crystalline states that provide kinetic pathways to metastable phases that are not accessible during near-equilibrium processing.

36 MATERIALS SCIENCE↗

RENDEZVOUS AND DOCKING TECHNIQUES

To implement the primary space mission of this decade – manned lunar exploration -- the operational assistance of rendezvous and docking is being considered as an alternative to possible problems in obtaining a boost vehicle capable of direct flight. This paper concentrates on the mechanization of the rendezvous and docking phase of such a lunar mission. As indicated in previous papers, rendezvous can take place in either an earth or lunar orbit (or on the lunar surface); it can involve the mating of stages, transfer of fuel, or the return of a shuttle to a “mother” ship; direct ascent or parking orbits can be used; the orbits can be circular or more general ellipses; finally, either or both of the spacecraft can participate in the rendezvous maneuvers. Since there is no intent here to discuss the pros and cons of each approach or to cover all possible mission profiles, guidance schemes and hardware configurations, one particular profile has been selected to display the significant features of most rendezvous missions. In many respects, rendezvous is less difficult than aircraft interception since the target is friendly and there is no severe time constraint. The latter feature allows considerable freedom of design especially with the great versatility of a human in the loop. This remains true even for a purely automatic mode. Perhaps the largest design problem concerns the selection of the mission to be implemented followed by the optimization or systems engineering of a mechanization from the multitude of possible schemes and techniques. A most important factor in the optimization is that of reliability and the redundancy, alternate or backup modes, etc., associated with the approach. While a primary system can be rather easily mechanized with modest equipment requirements, reliability considerations will result in additional hardware, tighter specifications and more safety factors (e. g., propellant margin). This is a very complex area and will only briefly be mentioned in the following discussion.

Rendezvous↗

Can Laws Be a Potential PET Image Texture Analysis Approach for Evaluation of Tumor Heterogeneity and Histopathological Characteristics in NSCLC?

We investigated the association between the textural features obtained from {sup 18}F-FDG images, metabolic parameters (SUVmax{sub ,} SUVmean, MTV, TLG), and tumor histopathological characteristics (stage and Ki-67 proliferation index) in non-small cell lung cancer (NSCLC). The FDG-PET images of 67 patients with NSCLC were evaluated. MATLAB technical computing language was employed in the extraction of 137 features by using first order statistics (FOS), gray-level co-occurrence matrix (GLCM), gray-level run length matrix (GLRLM), and Laws’ texture filters. Textural features and metabolic parameters were statistically analyzed in terms of good discrimination power between tumor stages, and selected features/parameters were used in the automatic classification by k-nearest neighbors (k-NN) and support vector machines (SVM). We showed that one textural feature (gray-level nonuniformity, GLN) obtained using GLRLM approach and nine textural features using Laws’ approach were successful in discriminating all tumor stages, unlike metabolic parameters. There were significant correlations between Ki-67 index and some of the textural features computed using Laws’ method (r = 0.6, p = 0.013). In terms of automatic classification of tumor stage, the accuracy was approximately 84% with k-NN classifier (k = 3) and SVM, using selected five features. Texture analysis of FDG-PET images has a potential to be an objective tool to assess tumor histopathological characteristics. The textural features obtained using Laws’ approach could be useful in the discrimination of tumor stage.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

STELLAR-MASS BLACK HOLE SPIN CONSTRAINTS FROM DISK REFLECTION AND CONTINUUM MODELING

Accretion disk reflection spectra, including broad iron emission lines, bear the imprints of the strong Doppler shifts and gravitational redshifts close to black holes. The extremity of these shifts depends on the proximity of the innermost stable circular orbit to the black hole, and that orbit is determined by the black hole spin parameter. Modeling relativistic spectral features, then, gives a means of estimating black hole spin. We report on the results of fits made to archival X-ray spectra of stellar-mass black holes and black hole candidates, selected for strong disk reflection features. Following recent work, these spectra were fit with reflection models and disk continuum emission models (where required) in which black hole spin is a free parameter. Although our results must be regarded as preliminary, we find evidence for a broad range of black hole spin parameters in our sample. The black holes with the most relativistic radio jets are found to have high spin parameters, though jets are observed in a black hole with a low spin parameter. For those sources with constrained binary system parameters, we examine the distribution of spin parameters versus black hole mass, binary mass ratio, and orbital period. We discuss the results within the context of black hole creation events, relativistic jet production, and efforts to probe the innermost relativistic regime around black holes.

79 ASTRONOMY AND ASTROPHYSICS↗

Machine learning-based surrogate modeling for data-driven optimization: a comparison of subset selection for regression techniques

Optimization of simulation-based or data-driven systems is a challenging task, which has attracted significant attention in the recent literature. A very efficient approach for optimizing systems without analytical expressions is through fitting surrogate models. Due to their increased flexibility, nonlinear interpolating functions, such as radial basis functions and Kriging, have been predominantly used as surrogates for data-driven optimization; however, these methods lead to complex nonconvex formulations. Alternatively, commonly used regression-based surrogates lead to simpler formulations, but they are less flexible and inaccurate if the form is not known a priori. In this work, we investigate the efficiency of subset selection regression techniques for developing surrogate functions that balance both accuracy and complexity. Subset selection creates sparse regression models by selecting only a subset of original features, which are linearly combined to generate a diverse set of surrogate models. Five different subset selection techniques are compared with commonly used nonlinear interpolating surrogate functions with respect to optimization solution accuracy, computation time, sampling requirements, and model sparsity. Furthermore, our results indicate that subset selection-based regression functions exhibit promising performance when the dimensionality is low, while interpolation performs better for higher dimensional problems.

42 ENGINEERING↗

Dependence of the Element Patterns of HYDROSTAR on Mutual Coupling

HYDROSTAR is a hybrid synthesis radiometer, intended for spaceborne applications, which employs a real aperture (waveguide stick antenna) for resolution along track and employs aperture synthesis to obtain resolution across track. This L-band system is an extension of the successful aircraft prototype, the Electronically Steered Thinned Array Radiometer (ESTAR). A proof-of-concept, full size system was constructed (45 wavelength in the synthesis direction) and subjected to extensive antenna pattern and associated microwave component measurements in an outdoor antenna test facility. HYDROSTAR employs a thinned array of 16 elements each 5.8 inches long in the along track dimension and spanning 9.5 inches across track. Each element is a narrow-wall shunt slot array with 36 slots. The polarization is linear (along-track). In the across track dimension, the antennas are deployed in a minimum redundancy array which has 90 independent baselines spaced in integer multiples of half a wavelength. The closest spacings used are for the first three elements at each end, which are spaced by only one-half wavelength. This study was intended to assess how closely each of 16 stick element patterns compare with their nominal values (as an individual, isolated radiator), when installed in their intended composite thinned array configuration. Extensive pattern measurements of the 16 elements that constitute the HYDROSTAR antenna subsystem were conducted to observe their relative features along the synthesis plane. Data was also collected for the mutual coupling between pairs of selected antennas. All antenna patterns had features different from that of an isolated element indicating some level of interaction among neighboring radiators. Those elements which had nearest neighbors at least five wavelengths away were located near the middle of the array. Their radiation patterns displayed sonic small, symmetric ripple across their full azimuth range. The patterns of elements that lie within 2.5 wavelengths of their neighbors showed stronger and asymmetric features. These are believed to be caused by mutual coupling among these structures. Evidence for this was seen when an antenna position was displaced by 0.05 wavelengths, Its pattern and those of its near neighbors were seen to change. Displacement within the plane of the array were observed to have different effects than displacements out-of-plane. A program of data analysis and theoretical development is in progress to provide a physical interpretation of the properties of these antenna patterns and to develop methods which can optimize the performance of this synthetic aperture imaging system. This includes compensation for pattern asymmetries and element position perturbation.

Weissman, David E.↗

ATP-Competitive Inhibitors of the Mammalian Target of Rapamycin: Design and Synthesis of Highly Potent and Selective Pyrazolopyrimidines

The mammalian target of rapamycin (mTOR), a central regulator of growth, survival, and metabolism, is a validated target for cancer therapy. Rapamycin and its analogues, allosteric inhibitors of mTOR, only partially inhibit one mTOR protein complex. ATP-competitive, global inhibitors of mTOR that have the potential for enhanced anticancer efficacy are described. Structural features leading to potency and selectivity were identified and refined leading to compounds with in vivo efficacy in tumor xenograft models.

59 BASIC BIOLOGICAL SCIENCES↗

Evaluating the effectiveness of program data features for guiding memory management

Recent trends have led to the adoption of larger and more complex memory systems, often with multiple tiers of memory performance within the same platform. To utilize complex memory systems efficiently, current data management strategies must be altered to map usage demands to the underlying hardware. Applications, as the generators of memory accesses, are well-suited to guide this process, but building and maintaining separate source code versions for different memory systems is not feasible in most cases. One potential solution is to employ automated program profiling and analysis to facilitate the production of application-based guidance. By attaching memory usage information to static or lightweight program features, compilers and runtime systems can generate fine-grained guidance without additional efforts from users or developers. Recent works have employed this approach with some success, but it is not clear which program features are most useful for guiding data management.This work evaluates the effectiveness of using different program data features to predict memory usage and guide memory management. It employs a custom set of simulation tools, based in the Intel® Pin framework, to collect and analyze the usage characteristics of application data associated with common program data features, such as allocation sites, types, and context. The results show that even relatively simple features, such as object size, are useful for selecting data with similar usage properties, but finer-grained features, such as the instructions that access a particular object, are often much more effective. Additionally, this work evaluates the performance of using different data features and different program inputs to guide data placement on a heterogeneous memory platform with a limited amount of high performance memory.

Effler, T. Chad↗