Publications

48 Results
Skip to search filters

LDRD 226360 Final Project Report: Simulated X-ray Diffraction and Machine Learning for Optimizing Dynamic Experiment Analysis

Ao, Tommy A.; Donohoe, Brendan D.; Martinez, Carianne M.; Knudson, Marcus D.; Montes de Oca Zapiain, David M.; Morgan, Dane M.; Rodriguez, Mark A.; Lane, James M.

This report is the final documentation for the one-year LDRD project 226360: Simulated X-ray Diffraction and Machine Learning for Optimizing Dynamic Experiment Analysis. As Sandia has successfully developed in-house X-ray diffraction tools for study of atomic structure in experiments, it has become increasingly important to develop computational analysis methods to support these experiments. When dynamically compressed lattices and orientations are not known a priori, the identification requires a cumbersome and sometimes intractable search of possible final states. These final states can include phase transition, deformation and mixed/evolving states. Our work consists of three parts: (1) development of an XRD simulation tool and use of traditional data science methods to match XRD patterns to experiments; (2) development of ML-based models capable of decomposing and identifying the lattice and orientation components of multicomponent experimental diffraction patterns; and (3) conducting experiments which showcase these new analysis tools in the study of phase transition mechanisms. Our target material has been cadmium sulfide, which exhibits complex orientation-dependent phase transformation mechanisms. In our current one-year LDRD, we have begun the analysis of high-quality c-axis CdS diffraction data from DCS and Thor experiments, which had until recently eluded orientation identification.

More Details

Super-Resolution Approaches in Three-Dimensions for Classification and Screening of Commercial-Off-The-Shelf Components

Polonsky, Andrew P.; Martinez, Carianne M.; Appleby, Catherine A.; Bernard, Sylvain R.; Griego, J.J.M.; Noell, Philip N.; Pathare, Priya R.

X-ray computed tomography is generally a primary step in characterization of defective electronic components, but is generally too slow to screen large lots of components. Super-resolution imaging approaches, in which higher-resolution data is inferred from lower-resolution images, have the potential to substantially reduce collection times for data volumes accessible via x-ray computed tomography. Here we seek to advance existing two-dimensional super-resolution approaches directly to three-dimensional computed tomography data. Multiple scan resolutions over a half order of magnitude of resolution were collected for four classes of commercial electronic components to serve as training data for a deep-learning, super-resolution network. A modular python framework for three-dimensional super-resolution of computed tomography data has been developed and trained over multiple classes of electronic components. Initial training and testing demonstrate the vast promise for these approaches, which have the potential for more than an order of magnitude reduction in collection time for electronic component screening.

More Details

Quantifying the unknown impact of segmentation uncertainty on image-based simulations

Nature Communications

Krygier, Michael K.; LaBonte, Tyler; Martinez, Carianne M.; Norris, Chance A.; Sharma, Krish; Collins, Lincoln; Mukherjee, Partha P.; Roberts, Scott A.

Image-based simulation, the use of 3D images to calculate physical quantities, relies on image segmentation for geometry creation. However, this process introduces image segmentation uncertainty because different segmentation tools (both manual and machine-learning-based) will each produce a unique and valid segmentation. First, we demonstrate that these variations propagate into the physics simulations, compromising the resulting physics quantities. Second, we propose a general framework for rapidly quantifying segmentation uncertainty. Through the creation and sampling of segmentation uncertainty probability maps, we systematically and objectively create uncertainty distributions of the physics quantities. We show that physics quantity uncertainty distributions can follow a Normal distribution, but, in more complicated physics simulations, the resulting uncertainty distribution can be surprisingly nontrivial. We establish that bounding segmentation uncertainty can fail in these nontrivial situations. While our work does not eliminate segmentation uncertainty, it improves simulation credibility by making visible the previously unrecognized segmentation uncertainty plaguing image-based simulation.

More Details

Credible, Automated Meshing of Images (CAMI)

Roberts, Scott A.; Donohoe, Brendan D.; Martinez, Carianne M.; Krygier, Michael K.; Hernandez-Sanchez, Bernadette A.; Foster, Collin W.; Collins, Lincoln; Greene, Benjamin G.; Noble, David R.; Norris, Chance A.; Potter, Kevin M.; Roberts, Christine C.; Neal, Kyle D.; Bernard, Sylvain R.; Schroeder, Benjamin B.; Trembacki, Bradley L.; LaBonte, Tyler L.; Sharma, Krish S.; Ganter, Tyler G.; Jones, Jessica E.; Smith, Matthew D.

Abstract not provided.

Confronting Domain Shift in Trained Neural Networks

Proceedings of Machine Learning Research

Martinez, Carianne M.; Najera-Flores, David A.; Brink, Adam R.; Quinn, D.D.; Chatzi, Eleni C.; Forrest, Stephanie F.

Neural networks (NNs) are known as universal function approximators and can interpolate nonlinear functions between observed data points. However, when the target domain for deployment shifts from the training domain and NNs must extrapolate, the results are notoriously poor. Prior work Martinez et al. (2019) has shown that NN uncertainty estimates can be used to correct binary predictions in shifted domains without retraining the model. We hypothesize that this approach can be extended to correct real-valued time series predictions. As an exemplar, we consider two mechanical systems with nonlinear dynamics. The first system consists of a spring-mass system where the stiffness changes abruptly, and the second is a real experimental system with a frictional joint that is an open challenge for structural dynamicists to model efficiently. Our experiments will test whether 1) NN uncertainty estimates can identify when the input domain has shifted from the training domain and 2) whether the information used to calculate uncertainty estimates can be used to correct the NN’s time series predictions. While the method as proposed did not significantly improve predictions, our results did show potential for modifications that could improve models’ predictions and play a role in structural health monitoring systems that directly impact public safety.

More Details

Segmentation certainty through uncertainty: Uncertainty-refined binary volumetric segmentation under multifactor domain shift

IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops

Martinez, Carianne M.; Potter, Kevin M.; Smith, Matthew D.; Donahue, Emily D.; Collins, Lincoln; Korbin, John P.; Roberts, Scott A.

Deep learning segmentation models are known to be sensitive to the scale, contrast, and distribution of pixel values when applied to Computed Tomography (CT) images. For material samples, scans are often obtained from a variety of scanning equipment and resolutions resulting in domain shift. The ability of segmentation models to generalize to examples from these shifted domains relies on how well the distribution of the training data represents the overall distribution of the target data. We present a method to overcome the challenges presented by domain shifts. Our results indicate that we can leverage a deep learning model trained on one domain to accurately segment similar materials at different resolutions by refining binary predictions using uncertainty quantification (UQ). We apply this technique to a set of unlabeled CT scans of woven composite materials with clear qualitative improvement of binary segmentations over the original deep learning predictions. In contrast to prior work, our technique enables refined segmentations without the expense of the additional training time and parameters associated with deep learning models used to address domain shift.

More Details
48 Results
48 Results