Alex Hagen

Alex Hagen, Ken Jarman, Jesse Ward et al.

Many measurements in the physical sciences can be cast as counting experiments, where the number of occurrences of a physical phenomenon informs the prevalence of the phenomenon's source. Often, detection of the physical phenomenon (termed signal) is difficult to distinguish from naturally occurring phenomena (termed background). In this case, the discrimination of signal events from background can be performed using classifiers, and they may range from simple, threshold-based classifiers to sophisticated neural networks. These classifiers are often trained and validated to obtain optimal accuracy, however we show that the optimal accuracy classifier does not generally coincide with a classifier that provides the lowest detection limit, nor the lowest quantification uncertainty. We present a derivation of the detection limit and quantification uncertainty in the classifier-based counting experiment case. We also present a novel abstention mechanism to minimize the detection limit or quantification uncertainty \emph{a posteriori}. We illustrate the method on two data sets from the physical sciences, discriminating Ar-37 and Ar-39 radioactive decay from non-radioactive events in a gas proportional counter, and discriminating neutrons from photons in an inorganic scintillator and report results therefrom.

Alex Hagen, Karl Pazdernik, Nicole LaHaye et al.

Measurement of uncertainty of predictions from machine learning methods is important across scientific domains and applications. We present, to our knowledge, the first such technique that quantifies the uncertainty of predictions from a classifier and accounts for both the classifier's belief and performance. We prove that our method provides an accurate estimate of the probability that the outputs of two neural networks are correct by showing an expected calibration error of less than 0.2% on a binary classifier, and less than 3% on a semantic segmentation network with extreme class imbalance. We empirically show that the uncertainty returned by our method is an accurate measurement of the probability that the classifier's prediction is correct and, therefore has broad utility in uncertainty propagation.

Alex Hagen, Shane Jackson, James Kahn et al.

Statistical testing is widespread and critical for a variety of scientific disciplines. The advent of machine learning and the increase of computing power has increased the interest in the analysis and statistical testing of multidimensional data. We extend the powerful Kolmogorov-Smirnov two sample test to a high dimensional form in a similar manner to Fasano (Fasano, 1987). We call our result the d-dimensional Kolmogorov-Smirnov test (ddKS) and provide three novel contributions therewith: we develop an analytical equation for the significance of a given ddKS score, we provide an algorithm for computation of ddKS on modern computing hardware that is of constant time complexity for small sample sizes and dimensions, and we provide two approximate calculations of ddKS: one that reduces the time complexity to linear at larger sample sizes, and another that reduces the time complexity to linear with increasing dimension. We perform power analysis of ddKS and its approximations on a corpus of datasets and compare to other common high dimensional two sample tests and distances: Hotelling's T^2 test and Kullback-Leibler divergence. Our ddKS test performs well for all datasets, dimensions, and sizes tested, whereas the other tests and distances fail to reject the null hypothesis on at least one dataset. We therefore conclude that ddKS is a powerful multidimensional two sample test for general use, and can be calculated in a fast and efficient manner using our parallel or approximate methods. Open source implementations of all methods described in this work are located at https://github.com/pnnl/ddks.

Alex Hagen, Shane Jackson

Detection of effects of the parameters of the synthetic process on the microstructure of materials is an important, yet elusive goal of materials science. We develop a method for detecting effects based on copula theory, high dimensional distribution distances, and permutational statistics to analyze a designed experiment synthesizing plutonium oxide from Pu(III) Oxalate. We detect effects of strike order and oxalic acid feed on the microstructure of the resulting plutonium oxide, which match the literature well. We also detect excess bivariate effects between the pairs of acid concentration, strike order and precipitation temperature.

Peter Lalor, Ayush Panigrahy, Alex Hagen

Training machine learning models for radioisotope identification using gamma spectroscopy remains an elusive challenge for many practical applications, largely stemming from the difficulty of acquiring and labeling large, diverse experimental datasets. Simulations can mitigate this challenge, but the accuracy of models trained on simulated data can deteriorate substantially when deployed to an out-of-distribution operational environment. In this study, we demonstrate that unsupervised domain adaptation (UDA) can improve the ability of a model trained on synthetic data to generalize to a new testing domain, provided unlabeled data from the target domain is available. Conventional supervised techniques are unable to utilize this data because the absence of isotope labels precludes defining a supervised classification loss. We compare a range of different UDA techniques, finding that feature alignment strategies, particularly via maximum mean discrepancy (MMD) minimization or domain-adversarial training, yield the most consistent improvement to testing scores. For instance, using a custom transformer-based neural network, we achieve a testing accuracy of $0.904 \pm 0.022$ on an experimental LaBr$_3$ test set after performing unsupervised feature alignment via MMD minimization, compared to $0.754 \pm 0.014$ before alignment. Overall, our results highlight the potential of using UDA to adapt a radioisotope classifier trained on synthetic data for real-world deployment.

Pedro Rodríguez Fernández, Christian Svinth, Alex Hagen

We present a method to improve the detection limit for radionuclides using spectroscopic radiation detectors and the arrival time of each detected radiation quantum. We enable this method using a neural network with an attention mechanism. We illustrate the method on the detection of Cesium release from a nuclear facility during an upset, and our method shows $2\times$ improvement over the traditional spectroscopic method. We hypothesize that our method achieves this performance increase by modulating its detection probability by the overall rate of probable detections, specifically by adapting detection thresholds based on temporal event distributions and local spectral features, and show evidence to this effect. We believe this method is applicable broadly and may be more successful for radionuclides with more complicated decay chains than Cesium; we also note that our method can generalize beyond the addition of arrival time and could integrate other data about each detection event, such as pulse quality, location in detector, or even combining the energy and time from detections in different detectors.

Marjolein Oostrom, Alex Hagen, Nicole LaHaye et al.

Understanding the relationship between the evolution of microstructures of irradiated LiAlO2 pellets and tritium diffusion, retention and release could improve predictions of tritium-producing burnable absorber rod performance. Given expert-labeled segmented images of irradiated and unirradiated pellets, we trained Deep Convolutional Neural Networks to segment images into defect, grain, and boundary classes. Qualitative microstructural information was calculated from these segmented images to facilitate the comparison of unirradiated and irradiated pellets. We tested modifications to improve the sensitivity of the model, including incorporating meta-data into the model and utilizing uncertainty quantification. The predicted segmentation was similar to the expert-labeled segmentation for most methods of microstructural qualification, including pixel proportion, defect area, and defect density. Overall, the high performance metrics for the best models for both irradiated and unirradiated images shows that utilizing neural network models is a viable alternative to expert-labeled images.

Peter Lalor, Henry Adams, Alex Hagen

Machine learning has the potential to improve the speed and reliability of radioisotope identification using gamma spectroscopy. However, meticulously labeling an experimental dataset for training is often prohibitively expensive, while training models purely on synthetic data is risky due to the domain gap between simulated and experimental measurements. In this research, we demonstrate that supervised domain adaptation can substantially improve the performance of radioisotope identification models by transferring knowledge between synthetic and experimental data domains. We consider two domain adaptation scenarios: (1) a simulation-to-simulation adaptation, where we perform multi-label proportion estimation using simulated high-purity germanium detectors, and (2) a simulation-to-experimental adaptation, where we perform multi-class, single-label classification using measured spectra from handheld lanthanum bromide (LaBr) and sodium iodide (NaI) detectors. We begin by pretraining a spectral classifier on synthetic data using a custom transformer-based neural network. After subsequent fine-tuning on just 64 labeled experimental spectra, we achieve a test accuracy of 96% in the sim-to-real scenario with a LaBr detector, far surpassing a synthetic-only baseline model (75%) and a model trained from scratch (80%) on the same 64 spectra. Furthermore, we demonstrate that domain-adapted models learn more human-interpretable features than experiment-only baseline models. Overall, our results highlight the potential for supervised domain adaptation techniques to bridge the sim-to-real gap in radioisotope identification, enabling the development of accurate and explainable classifiers even in real-world scenarios where access to experimental data is limited.

Alex Hagen, Eric Church, Jan Strube et al.

Measurements in Liquid Argon Time Projection Chamber (LArTPC) neutrino detectors, such as the MicroBooNE detector at Fermilab, feature large, high fidelity event images. Deep learning techniques have been extremely successful in classification tasks of photographs, but their application to LArTPC event images is challenging, due to the large size of the events. Events in these detectors are typically two orders of magnitude larger than images found in classical challenges, like recognition of handwritten digits contained in the MNIST database or object recognition in the ImageNet database. Ideally, training would occur on many instances of the entire event data, instead of many instances of cropped regions of interest from the event data. However, such efforts lead to extremely long training cycles, which slow down the exploration of new network architectures and hyperparameter scans to improve the classification performance. We present studies of scaling a LArTPC classification problem on multiple architectures, spanning multiple nodes. The studies are carried out on simulated events in the MicroBooNE detector. We emphasize that it is beyond the scope of this study to optimize networks or extract the physics from any results here. Institutional computing at Pacific Northwest National Laboratory and the SummitDev machine at Oak Ridge National Laboratory's Leadership Computing Facility have been used. To our knowledge, this is the first use of state-of-the-art Convolutional Neural Networks for particle physics and their attendant compute techniques onto the DOE Leadership Class Facilities. We expect benefits to accrue particularly to the Deep Underground Neutrino Experiment (DUNE) LArTPC program, the flagship US High Energy Physics (HEP) program for the coming decades.

Brian Thomas, Tim Jenness, Frossie Economou et al.

The Flexible Image Transport System (FITS) standard has been a great boon to astronomy, allowing observatories, scientists and the public to exchange astronomical information easily. The FITS standard, however, is showing its age. Developed in the late 1970s, the FITS authors made a number of implementation choices that, while common at the time, are now seen to limit its utility with modern data. The authors of the FITS standard could not anticipate the challenges which we are facing today in astronomical computing. Difficulties we now face include, but are not limited to, addressing the need to handle an expanded range of specialized data product types (data models), being more conducive to the networked exchange and storage of data, handling very large datasets, and capturing significantly more complex metadata and data relationships. There are members of the community today who find some or all of these limitations unworkable, and have decided to move ahead with storing data in other formats. If this fragmentation continues, we risk abandoning the advantages of broad interoperability, and ready archivability, that the FITS format provides for astronomy. In this paper we detail some selected important problems which exist within the FITS standard today. These problems may provide insight into deeper underlying issues which reside in the format and we provide a discussion of some lessons learned. It is not our intention here to prescribe specific remedies to these issues; rather, it is to call attention of the FITS and greater astronomical computing communities to these problems in the hope that it will spur action to address them.