Search NASA⌕ Search

SEARCH · Search NASA

Results for “statistical learning”

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 1,711 records · Page 95

Modification of Eccentric Gaze-Holding

Clear vision and accurate localization of objects in the environment are prerequisites for reliable performance of motor tasks. Space flight confronts the crewmember with a stimulus rearrangement that requires adaptation to function effectively with the new requirements of altered spatial orientation and motor coordination. Adaptation and motor learning driven by the effects of cerebellar disorders may share some of the same demands that face our astronauts. One measure of spatial localization shared by the astronauts and those suffering from cerebellar disorders that is easily quantified, and for which a neurobiological substrate has been identified, is the control of the angle of gaze (the "line of sight"). The disturbances of gaze control that have been documented to occur in astronauts and cosmonauts, both in-flight and postflight, can be directly related to changes in the extrinsic gravitational environment and intrinsic proprioceptive mechanisms thus, lending themselves to description by simple non-linear statistical models. Because of the necessity of developing robust normal response populations and normative populations against which abnormal responses can be evaluated, the basic models can be formulated using normal, non-astronaut test subjects and subsequently extended using centrifugation techniques to alter the gravitational and proprioceptive environment of these subjects. Further tests and extensions of the models can be made by studying abnormalities of gaze control in patients with cerebellar disease. A series of investigations were conducted in which a total of 62 subjects were tested to: (1) Define eccentric gaze-holding parameters in a normative population, and (2) explore the effects of linear acceleration on gaze-holding parameters. For these studies gaze-holding was evaluated with the subjects seated upright (the normative values), rolled 45 degrees to both the left and right, or pitched back 30 and 90 degrees. In a separate study the further effects of acceleration on gaze stability was examined during centrifugation (+2 G (sub x) and +2 G (sub z) using a total of 23 subjects. In all of our investigations eccentric gaze-holding was established by having the subjects acquire an eccentric target (+/-30 degrees horizontal, +/- 15 degrees vertical) that was flashed for 750 msec in an otherwise dark room. Subjects were instructed to hold gaze on the remembered position of the flashed target for 20 sec. Immediately following the 20 sec period, subjects were cued to return to the remembered center position and to hold gaze there for an additional 20 sec. Following this 20 sec period the center target was briefly flashed and the subject made any corrective eye movement back to the true center position. Conventionally, the ability to hold eccentric gaze is estimated by fitting the natural log of centripetal eye drifts by linear regression and calculating the time constant (G) of these slow phases of "gaze-evoked nystagmus". However, because our normative subjects sometimes showed essentially no drift (tau (sub c) = m), statistical estimation and inference on the effect of target direction was performed on values of the decay constant theta = 1/(tau (sub c)) which we found was well modeled by a gamma distribution. Subjects showed substantial variance of their eye drifts, which were centrifugal in approximately 20 % of cases, and > 40% for down gaze. Using the ensuing estimated gamma distributions, we were able to conclude that rightward and leftward gaze holding were not significantly different, but that upward gaze holding was significantly worse than downward (p<0.05). We also concluded that vertical gaze holding was significantly worse than horizontal (p<0.05). In the case of left and right roll, we found that both had a similar improvement to horizontal gaze holding (p<0.05), but didn't have a significant effect on vertical gaze holding. For pitch tilts, both tilt angles significantly decreased gaze-holding ility in all directions (p<0.05). Finally, we found that hyper-g centrifugation significantly decreased gaze holding ability in the vertical plane. The main findings of this study are as follows: (1) vertical gaze-holding is less stable than horizontal, (2) gaze-holding to upward targets is less stable than to downward targets, (3) tilt affects gaze holding, and (4) hyper-g affects gaze holding. This difference between horizontal and vertical gaze-holding may be ascribed to separate components of the velocity-to-position neural integrator for eye movements, and to differences in orbital mechanics. The differences between upward and downward gaze-holding may be ascribed to an inherent vertical imbalance in the vestibular system. Because whole body tilt and hyper-g affects gaze-holding, it is implied that the otolith organs have direct connections to the neural integrator and further studies of astronaut gaze-holding are warranted. Our statistical method for representing the range of normal eccentric gaze stability can be readily applied to normals who maybe exposed to environments which may modify the central integrator and require monitoring, and to evaluate patients with gaze-evoked nystagmus by comparing to the above established normative criteria.

Reschke, M. F.↗

Real-Time event reconstruction for Nuclear Physics Experiments using Artificial Intelligence

Charged track reconstruction is a critical task in nuclear physics experiments, enabling the identification and analysis of particles produced in high-energy collisions. Machine learning (ML) has emerged as a powerful tool for this purpose, addressing the challenges posed by complex detector geometries, high event multiplicities, and noisy data. Traditional methods rely on pattern recognition algorithms like the Kalman filter, but ML techniques, such as neural networks, graph neural networks (GNNs), and recurrent neural networks (RNNs), offer improved accuracy and scalability. By learning from simulated and real detector data, ML models can identify and classify tracks, predict trajectories, and handle ambiguities caused by overlapping or missing hits. Moreover, ML-based approaches can process data in near-real-time, enhancing the efficiency of experiments at large-scale facilities like the Large Hadron Collider (LHC) and Jefferson Lab (JLAB). As detector technologies and computational resources evolve, ML-driven charged track reconstruction continues to push the boundaries of precision and discovery in nuclear physics. In these proceedings, we highlight advancements in charged track identification leveraging Artificial Intelligence within the CLAS12 detector, achieving a notable enhancement in experimental statistics compared to traditional methods. Additionally, we showcase real-time event reconstruction capabilities, including the inference of charged particle properties, such as momentum, direction, and species identification, at speeds matching data acquisition rates. These innovations enable the extraction of physics observables directly from the experiment in real-time.

Gavalian, Gagik (ORCID:0000000267385457)↗

A Machine Learning Framework for Modeling Ensemble Properties of Atomically Disordered Materials

Atomic disorder can strongly influence material properties such as charge transport, optical response, and catalytic activity. However, efficiently modeling these disorder effects remains challenging for first-principles methods due to the cost of sampling large configurational spaces and computing complex physical quantities. Recent advances of machine learning techniques, particularly graph neural networks (GNNs), has enabled the efficient and accurate predictions of complex material properties, offering promising tools for studying disordered systems. In this work, we present a general machine-learning-assisted computational framework that integrates equivariant GNNs with Monte Carlo simulations to compute the thermodynamic and ensemble-averaged functional properties of disordered materials. Using the surface-termination-disordered MXene monolayer Ti 3 C 2 T 2–x as a representative system, we find that electrical conductivity exhibits an emergent peak near the order–disorder phase transition temperature due to the interplay between electron scattering and doping. In contrast, optical conductivity remains largely insensitive to local atomic disorder and reflects the global surface chemical composition. These results highlight the role of atomic disorder in affecting material properties and demonstrate the potential of our approach for statistically modeling disorder effects in a wide range of materials such as high-entropy alloys and spin liquids.

MXene↗

Pathway-based analyses of gene expression profiles at low doses of ionizing radiation

Radiation exposure poses a significant threat to human health. Emerging research indicates that even low-dose radiation once believed to be safe, may have harmful effects. This perception has spurred a growing interest in investigating the potential risks associated with low-dose radiation exposure across various scenarios. To comprehensively explore the health consequences of low-dose radiation, our study employs a robust statistical framework that examines whether specific groups of genes, belonging to known pathways, exhibit coordinated expression patterns that align with the radiation levels. Notably, our findings reveal the existence of intricate yet consistent signatures that reflect the molecular response to radiation exposure, distinguishing between low-dose and high-dose radiation. Moreover, we leverage a pathway-constrained variational autoencoder to capture the nonlinear interactions within gene expression data. By comparing these two analytical approaches, our study aims to gain valuable insights into the impact of low-dose radiation on gene expression patterns, identify pathways that are differentially affected, and harness the potential of machine learning to uncover hidden activity within biological networks. This comparative analysis contributes to a deeper understanding of the molecular consequences of low-dose radiation exposure.

63 RADIATION, THERMAL, AND OTHER ENVIRON. POLLUTAN↗

Parameterization of Vertical Cloud Distribution from C3M and MERRA Data Using ML Method

Clouds play a key role in regulating the hydrological cycle and the Earth's radiative energy budget. However, global climate models (GCMs) with a horizontal grid spacing on the order of 100 km have limitations in representing sub-grid cloud dynamics with spatial scales on the order of 1 km, leading to potential uncertainties in cloud radiative feedback on the global scale. In our research, we will leverage the capabilities of Deep Machine Learning (DML) methods to construct parameterizations of sub-grid volumetric cloud fraction (VCF), which is the frequency of occurrence on a grid volume accumulated in the horizontal and vertical directions. Our investigation delves into the intricate relationship between VCF obtained from the NASA CALIPSO-CloudSat-CERES-MODIS (CCCM) satellite observation data and 3-D MERRA-2 reanalysis meteorological profiling data (e.g., wind, relative humidity, temperature). Through a comprehensive one-year data training utilizing the Sequence to Sequence DML method, we have successfully disentangled the complicated cloud formation dynamics across diverse meteorological conditions through a day-to-day analysis framework. Preliminary findings reveal promising statistical agreements in geographical and vertical distributions and seasonal variations of volumetric cloud fraction between ML prediction and satellite measurements. These results underscore the aptitude of our DML model to discern underlying cloud physical processes and accurately represent sub-grid cloud formation dynamics. Additionally, we have also employed trained neural network to analyze uncertainties arising from errors in meteorological data, further enhancing the robustness of our VCF parameterization.

Shan Zeng↗

Large-scale tearing-mode hazard function analysis with standard matched equilibrium reconstructions

The association between features from standard tokamak equilibrium reconstructions and the onset of n = 1 tearing modes (TMs) is analyzed at scale. The TM onset rate is directly modeled with a ‘hazard’ function which gives the expected number of onsets (per unit time spent) in a given equilibrium parameter region. In particular the different statistical modeling performance achieved for magnetics-only reconstructions and motional Stark effect (MSE) enhanced reconstructions is studied. It is observed that a better hazard model for the TM onset rate can be built with the MSE-enhanced equilibria compared to the matched magnetics-only situation. This advantage disappears if internal profile details are withheld from the matched analysis. Plausibility of the hazard function is further demonstrated with visualizations of global trends in the operational space, and time-traces from specific tokamak discharges. As a result, TMs typically degrade tokamak plasma performance and may lead to plasma termination, motivating this statistical study.

equilibrium↗

SU-E-T-580: On the Significance of Model Based Dosimetry for Breast and Head and Neck 192Ir HDR Brachytherapy

Purpose: To assess the dosimetric and radiobiological differences between TG43-based and model-based dosimetry in the treatment planning of {sup 192}Ir HDR brachytherapy for breast and head and neck cancer. Methods: Two cohorts of 57 Accelerated Partial Breast Irradiation (APBI) and 22 head and neck (H&N) patients with oral cavity carcinoma were studied. Dosimetry for the treatment plans was performed using the TG43 algorithm of the Oncentra Brachy v4.4 treatment planning system (TPS). Corresponding Monte Carlo (MC) simulations were performed using MCNP6 with input files automatically prepared by the BrachyGuide software tool from DICOM RT plan data. TG43 and MC data were compared in terms of % dose differences, Dose Volume Histograms (DVHs) and related indices of clinical interest for the Planning Target Volume (PTV) and the Organs-At-Risk (OARs). A radiobiological analysis was also performed using the Equivalent Uniform Dose (EUD), mean survival fraction (S) and Tumor Control Probability (TCP) for the PTV, and the Normal Tissue Control Probability (N TCP) and the generalized EUD (gEUD) for the OARs. Significance testing of the observed differences performed using the Wilcoxon paired sample test. Results: Differences between TG43 and MC DVH indices, associated with the increased corresponding local % dose differences observed, were statistically significant. This is mainly attributed to their consistency however, since TG43 agrees closely with MC for the majority of DVH and radiobiological parameters in both patient cohorts. Differences varied considerably among patients only for the ipsilateral lung and ribs in the APBI cohort, with a strong correlation to target location. Conclusion: While the consistency and magnitude of differences in the majority of clinically relevant DVH indices imply that no change is needed in the treatment planning practice, individualized dosimetry improves accuracy and addresses instances of inter-patient variability observed. Research co-financed by the ESF and Greek funds through the Operational Program Education and Lifelong Learning Investing in Knowledge Society of the NSRF. Research Funding Program Aristeia. Nucletron, an Elekta company (Veenendaal, The Netherlands) is gratefully acknowledged for providing Oncentra Brachy v4.4 for research purposes.

61 RADIATION PROTECTION AND DOSIMETRY↗

Prediction of Solar Radiation on Building Rooftops: A Data-Mining Approach

Solar energy technologies offer a clean, renewable, and domestic energy source, and are essential components of a sustainable energy future. The accurate measurement of solar radiation data is essential for optimum site selection of future distributed solar power plants as well as sizing photovoltaic systems. However, solar radiation data are not readily available because measured sequences of radiation values are obtained for a few locations in a country. When the data are available, they are usually at different time periods and spatial scale. The availability of solar radiation data at hourly or daily time scale will enhance the integration of solar energy into electricity generation and promote a sustainable energy future. The ability to generate approximate solar radiation values is often the only practical way to obtain radiation data at hourly or daily time scale. As a result, several models have been developed for estimating solar radiation values based on analytical, numerical simulation, and statistical approaches. However, these models have inherent challenges. We will discuss some of those challenges in this paper. To enhance the prediction of solar radiation values, a novel approach is presented for estimating solar radiation values using support vector machine technique. The approach accounts for unique characteristics that influence solar radiation values. The preliminary results obtained offer useful insights for model enhancements.

machine learning↗

Nuclear Forensics for Irradiated Materials using Kernel Regression Models

The ultimate purpose of nuclear forensics is to locate the origin of nuclear and other radioactive materials out of regulatory control used in nuclear security events such as classical forensics. For this purpose, some specific characteristics of interdicted materials (referred to as 'signatures') should be defined to compare the known sample database. In some cases, involved with irradiated material, the signatures could be specific amount of fission product inventory depending on reactor operational histories such as reactor type, initial enrichment, burn up rate of nuclear fuel and cooling period after irradiation in the reactor. In order to make inference model for irradiated material for nuclear forensics, we suggested 3 ways as follows; (i) analyzing ORIGEN produced nuclear substances for the fission product inventory, (ii) classifying the range of reactor type, (iii) predicting the operational history from the fission product inventory. In terms of nuclear substance analysis, the construction of spent fuel database has been internationally carried out since ITWG (Nuclear Smuggling International Technical Working Group) was established in 1995. However, enough sample data cannot be obtained in Korea to determine reactor type and operational histories because analyzing nuclear substances have been limited due to the international regulation. Thus, this study does not mention how to obtain mass or ratios of nuclides for the spent fuel. We just have generated simulation sample data as a fission product inventory using ORIGEN to construct spent fuel database. When unidentified radioactive materials are discovered, it is assumed that mass or ratios of the nuclides are analyzed by certain methods which are available in Korea as well as other countries. Then, a reactor type is determined, and finally, regression model can locate the origin or source of the radioactive materials by inferring operational histories using the fission product inventory. There are some studies on the classification of reactor type using statistical methods. As a part of predicting the operational history, this paper focuses on the development of spent fuel inference algorithm for nuclear forensics to estimate reactor operational histories that reflect characteristics of a reactor from the fission product inventory. The spent fuel inference algorithm for nuclear forensics has been developed on the basis of Kernel Regression (KR) which can conditionally generate the estimated output data by giving weight depending on geometric distance between sample data and input data. Also the KR which comprises Auto-Associative KR and Inferential KR have some benefits that can estimate only using data without assumption for specific physical model and re-learning process. Auto-associative KR (AAKR) was used to sort out outliers of input data and Inferential KR (IKR) was carried out to predict the operational histories for spent fuel. All of regression algorithms were performed in R programming tool. From the validation of both AAKR and IKR, it is expected that kernel regression has enough accuracy to be applied to nuclear forensics for spent fuel. Once the nuclear substance is measured to fill the fission product inventory with a certain reactor type, this kernel regression algorithm would predict the operational histories of spent fuel with high accuracy. Finally, this information is expected to be useful to find the origin of irradiated materials. (authors)

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Exploring Saccharomycotina Yeast Ecology Through an Ecological Ontology Framework

Yeasts in the subphylum Saccharomycotina are found across the globe in disparate ecosystems. A major aim of yeast research is to understand the diversity and evolution of ecological traits, such as carbon metabolic breadth, insect association, and cactophily. This includes studying aspects of ecological traits like genetic architecture or association with other phenotypic traits. Genomic resources in the Saccharomycotina have grown rapidly. Ecological data, however, are still limited for many species, especially those only known from species descriptions where usually only a limited number of strains are studied. Moreover, ecological information is recorded in natural language format limiting high throughput computational analysis. To address these limitations, we developed an ontological framework for the analysis of yeast ecology. A total of 1,088 yeast strains were added to the Ontology of Yeast Environments (OYE) and analyzed in a machine-learning framework to connect genotype to ecology. This framework is flexible and can be extended to additional isolates, species, or environmental sequencing data. Widespread adoption of OYE would greatly aid the study of macroecology in the Saccharomycotina subphylum.

59 BASIC BIOLOGICAL SCIENCES↗

Correspondence between neuroevolution and gradient descent

Abstract We show analytically that training a neural network by conditioned stochastic mutation or neuroevolution of its weights is equivalent, in the limit of small mutations, to gradient descent on the loss function in the presence of Gaussian white noise. Averaged over independent realizations of the learning process, neuroevolution is equivalent to gradient descent on the loss function. We use numerical simulation to show that this correspondence can be observed for finite mutations, for shallow and deep neural networks. Our results provide a connection between two families of neural-network training methods that are usually considered to be fundamentally different.

97 MATHEMATICS AND COMPUTING↗

Automatic detection of low surface brightness galaxies from Sloan Digital Sky Survey images

ABSTRACT Low surface brightness (LSB) galaxies are galaxies with central surface brightness fainter than the night sky. Due to the faint nature of LSB galaxies and the comparable sky background, it is difficult to search LSB galaxies automatically and efficiently from large sky survey. In this study, we established the low surface brightness galaxies autodetect (LSBG-AD) model, which is a data-driven model for end-to-end detection of LSB galaxies from Sloan Digital Sky Survey (SDSS) images. Object-detection techniques based on deep learning are applied to the SDSS field images to identify LSB galaxies and estimate their coordinates at the same time. Applying LSBG-AD to 1120 SDSS images, we detected 1197 LSB galaxy candidates, of which 1081 samples are already known and 116 samples are newly found candidates. The B-band central surface brightness of the candidates searched by the model ranges from 22 to 24 mag arcsec−2, quite consistent with the surface brightness distribution of the standard sample. A total of 96.46 per cent of LSB galaxy candidates have an axial ratio (b/a) greater than 0.3, and 92.04 per cent of them have $fracDev\_r$ < 0.4, which is also consistent with the standard sample. The results show that the LSBG-AD model learns the features of LSB galaxies of the training samples well, and can be used to search LSB galaxies without using photometric parameters. Next, this method will be used to develop efficient algorithms to detect LSB galaxies from massive images of the next-generation observatories.

79 ASTRONOMY AND ASTROPHYSICS↗

SU-E-J-107: Supervised Learning Model of Aligned Collagen for Human Breast Carcinoma Prognosis

Purpose: Our goal is to develop and apply a set of optical and computational tools to enable large-scale investigations of the interaction between collagen and tumor cells. Methods: We have built a novel imaging system for automating the capture of whole-slide second harmonic generation (SHG) images of collagen in registry with bright field (BF) images of hematoxylin and eosin stained tissue. To analyze our images, we have integrated a suite of supervised learning tools that semi-automatically model and score collagen interactions with tumor cells via a variety of metrics, a method we call Electronic Tumor Associated Collagen Signatures (eTACS). This group of tools first segments regions of epithelial cells and collagen fibers from BF and SHG images respectively. We then associate fibers with groups of epithelial cells and finally compute features based on the angle of interaction and density of the collagen surrounding the epithelial cell clusters. These features are then processed with a support vector machine to separate cancer patients into high and low risk groups. Results: We validated our model by showing that eTACS produces classifications that have statistically significant correlation with manual classifications. In addition, our system generated classification scores that accurately predicted breast cancer patient survival in a cohort of 196 patients. Feature rank analysis revealed that TACS positive fibers are more well aligned with each other, generally lower density, and terminate within or near groups of epithelial cells. Conclusion: We are working to apply our model to predict survival in larger cohorts of breast cancer patients with a diversity of breast cancer types, predict response to treatments such as COX2 inhibitors, and to study collagen architecture changes in other cancer types. In the future, our system may be used to provide metastatic potential information to cancer patients to augment existing clinical assays.

60 APPLIED LIFE SCIENCES↗

PylotDB - A Database Management, Graphing, and Analysis Tool Written in Python

PylotDB, written completely in Python, provides a user interface (UI) with which to interact with, analyze, graph data from, and manage open source databases such as MySQL. The UI mitigates the user having to know in-depth knowledge of the database application programming interface (API). PylotDB allows the user to generate various kinds of plots from user-selected data; generate statistical information on text as well as numerical fields; backup and restore databases; compare database tables across different databases as well as across different servers; extract information from any field to create new fields; generate, edit, and delete databases, tables, and fields; generate or read into a table CSV data; and similar operations. Since much of the database information is brought under control of the Python computer language, PylotDB is not intended for huge databases for which MySQL and Oracle, for example, are better suited. PylotDB is better suited for smaller databases that might be typically needed in a small research group situation. PylotDB can also be used as a learning tool for database applications in general.

Barnette, DanielW.↗

Understanding the Physics Representation of Deep Learning Models in Environmental Applications

Deep learning (DL) models have been popular in earth and environmental modeling and analysis, which exhibit huge potential in capturing and reconstructing the non-linearity of relevant environmental processes. They are extensively used as analytical tools or emulators for multiple domains (atmosphere, land surface, ocean, and biogeochemistry). Despite their success, their internal working mechanism remains largely unknown. Such a lack of knowledge hinders the identification of physically consistent models that are fully adaptive to non-stationary climate, as well as the development of physics-informed machine learning such as physics-informed neural network (PINN). To establish preliminary knowledge and framework of such physics representation evaluation, this project focuses on an improved understanding of DL models in the environmental applications. DL models are increasingly applied to environmental modeling and prediction. However, they have been evaluated mostly from a performance perspective, and there is a gap in understanding how they represent the known physics internally. Such knowledge is especially critical when applying DL models under climate change conditions, where new inputs are likely outside the ranges of the training datasets. In this project, we reveal how the known physical processes are represented within DL models from both statistical and mechanistic perspectives. Leveraging the traditional model evaluations that focus more on the accuracies of predictions, we establish a framework that examines both the accuracy and physics representation of DL models. This analysis framework can identify DL models that make the correct predictions based on correct physics, thus enhancing the existing explainable artificial intelligence (explainable-AI) portfolio. It lays a foundation for developing novel metrics to evaluate the emerging DL models in environmental applications. This knowledge also informs the development of physics-informed DL models by revealing the direct connections between the known physical processes and specific model components or structures.

54 ENVIRONMENTAL SCIENCES↗

Development of an inertial confinement fusion platform to study charged particle-producing nuclear reactions relevant to nuclear astrophysics

This paper describes the development of a platform to study astrophysically relevant nuclear reactions using inertial-confinement fusion implosions on the OMEGA and NIF laser facilities, with a particular focus on optimizing the implosions to study charged-particle-producing reactions. Primary requirements on the platform are high yield, for high statistics in the fusion product measurements, combined with low areal density, to allow the charged fusion products to escape. This is optimally achieved with direct-drive exploding pusher implosions using thin-glass-shell capsules. Mitigation strategies to eliminate a possible target sheath potential which would accelerate the emitted ions are discussed. The potential impact of kinetic effects on the implosions is also considered. The platform is initially employed to study the complementary T(t,2n)??, T(3He,np)?? and 3He(3He,2p)?? reactions. Proof-of-principle results from the first experiments demonstrating the ability to accurately measure the energy and yields of charged particles are presented. Lessons learned from these experiments will be used in studies of other reactions. The goals are to explore thermonuclear reaction rates and fundamental nuclear physics in stellar-like plasma environments, and to push this new frontier of nuclear astrophysics into unique regimes not reachable through existing platforms, with thermal ion velocity distributions, plasma screening and low reactant energies.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Fast Query-Optimized Kernel-Machine Classification

A recently developed algorithm performs kernel-machine classification via incremental approximate nearest support vectors. The algorithm implements support-vector machines (SVMs) at speeds 10 to 100 times those attainable by use of conventional SVM algorithms. The algorithm offers potential benefits for classification of images, recognition of speech, recognition of handwriting, and diverse other applications in which there are requirements to discern patterns in large sets of data. SVMs constitute a subset of kernel machines (KMs), which have become popular as models for machine learning and, more specifically, for automated classification of input data on the basis of labeled training data. While similar in many ways to k-nearest-neighbors (k-NN) models and artificial neural networks (ANNs), SVMs tend to be more accurate. Using representations that scale only linearly in the numbers of training examples, while exploring nonlinear (kernelized) feature spaces that are exponentially larger than the original input dimensionality, KMs elegantly and practically overcome the classic curse of dimensionality. However, the price that one must pay for the power of KMs is that query-time complexity scales linearly with the number of training examples, making KMs often orders of magnitude more computationally expensive than are ANNs, decision trees, and other popular machine learning alternatives. The present algorithm treats an SVM classifier as a special form of a k-NN. The algorithm is based partly on an empirical observation that one can often achieve the same classification as that of an exact KM by using only small fraction of the nearest support vectors (SVs) of a query. The exact KM output is a weighted sum over the kernel values between the query and the SVs. In this algorithm, the KM output is approximated with a k-NN classifier, the output of which is a weighted sum only over the kernel values involving k selected SVs. Before query time, there are gathered statistics about how misleading the output of the k-NN model can be, relative to the outputs of the exact KM for a representative set of examples, for each possible k from 1 to the total number of SVs. From these statistics, there are derived upper and lower thresholds for each step k. These thresholds identify output levels for which the particular variant of the k-NN model already leans so strongly positively or negatively that a reversal in sign is unlikely, given the weaker SV neighbors still remaining. At query time, the partial output of each query is incrementally updated, stopping as soon as it exceeds the predetermined statistical thresholds of the current step. For an easy query, stopping can occur as early as step k = 1. For more difficult queries, stopping might not occur until nearly all SVs are touched. A key empirical observation is that this approach can tolerate very approximate nearest-neighbor orderings. In experiments, SVs and queries were projected to a subspace comprising the top few principal- component dimensions and neighbor orderings were computed in that subspace. This approach ensured that the overhead of the nearest-neighbor computations was insignificant, relative to that of the exact KM computation.

Mazzoni, Dominic↗

Exploring physics of ferroelectric domain walls via Bayesian analysis of atomically resolved STEM data

The physics of ferroelectric domain walls is explored using the Bayesian inference analysis of atomically resolved STEM data. We demonstrate that domain wall profile shapes are ultimately sensitive to the nature of the order parameter in the material, including the functional form of Ginzburg-Landau-Devonshire expansion, and numerical value of the corresponding parameters. The preexisting materials knowledge naturally folds in the Bayesian framework in the form of prior distributions, with the different order parameters forming competing (or hierarchical) models. Here, we explore the physics of the ferroelectric domain walls in BiFeO 3 using this method, and derive the posterior estimates of relevant parameters. More generally, this inference approach both allows learning materials physics from experimental data with associated uncertainty quantification, and establishing guidelines for instrumental development answering questions on what resolution and information limits are necessary for reliable observation of specific physical mechanisms of interest.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗