Search

We address the problem of predicting a target ordinal variable based on observable features consisting of functional profiles. This problem is crucial, especially in decision-making driven by sensor systems, when the goal is to assess an ordinal variable such as the degree of deterioration, quality level, or risk stage of a process, starting from functional data observed via sensors. We purposely introduce a novel approach called functional-ordinal Canonical Correlation Analysis (foCCA), which is based on a functional data analysis approach. FoCCA allows for dimensionality reduction of observable features while maximizing their ability to differentiate between consecutive levels of an ordinal target variable. Unlike existing methods for supervised learning from functional data, foCCA fully incorporates the ordinal nature of the target variable. This enables the model to capture and represent the relative dissimilarities between consecutive levels of the ordinal target, while also explaining these differences through the functional features. Extensive simulations demonstrate that foCCA outperforms current state-of-the-art methods in terms of prediction accuracy in the reduced feature space. A case study involving the prediction of antigen concentration levels from optical biosensor signals further confirms the superior performance of foCCA, showcasing both improved predictive power and enhanced interpretability compared to competing approaches.

This expository paper introduces a simplified approach to image-based quality inspection in manufacturing using OpenAI's CLIP (Contrastive Language-Image Pretraining) model adapted for few-shot learning. While CLIP has demonstrated impressive capabilities in general computer vision tasks, its direct application to manufacturing inspection presents challenges due to the domain gap between its training data and industrial applications. We evaluate CLIP's effectiveness through five case studies: metallic pan surface inspection, 3D printing extrusion profile analysis, stochastic textured surface evaluation, automotive assembly inspection, and microstructure image classification. Our results show that CLIP can achieve high classification accuracy with relatively small learning sets (50-100 examples per class) for single-component and texture-based applications. However, the performance degrades with complex multi-component scenes. We provide a practical implementation framework that enables quality engineers to quickly assess CLIP's suitability for their specific applications before pursuing more complex solutions. This work establishes CLIP-based few-shot learning as an effective baseline approach that balances implementation simplicity with robust performance, demonstrated in several manufacturing quality control applications., Comment: 31 pages, 13 figures

Over the last five years, McGill University's Office of Science Education (OSE) has partnered with faculty members from the Department of Physics to form an education working group with the aim of charting the progression of students' conceptual understanding of uncertainties across their undergraduate degree. The research conducted by this group seeks to provide further insight into both the experimental skill set that students gain through undergraduate laboratory courses and how the department could address noticeable gaps in student understanding. In this paper, we evaluate the conceptual understanding of uncertainty using the Concise Data Processing Assessment (CDPA) instrument. First, we characterize the physics laboratory curriculum at McGill University by evaluating the evolution of CDPA scores across consecutive laboratory courses, and further propose the utilization of this tool for identifying gaps in student understanding. Following the analysis of student responses (N=2023), we specifically investigate data collected in second-year courses to better diagnose what student errors can tell us about common misconceptions in experimental physics. This more in-depth research focuses on data collected from students at the beginning and the end of their first full year of experimental laboratory courses, consisting of two consecutive laboratory courses that build on each other. By the end of the second course, students have engaged with all the material covered in the CDPA test. Interestingly, there have been no changes in CDPA total scores throughout the COVID-19 pandemic. We notice a marked upward shift in student understanding; however, the results indicate that a significant portion of students continue to struggle with uncertainties, basic data analysis, and curve fitting., Comment: 16 pages, 5 figures, 1 table

As prompt engineering research rapidly evolves, evaluations beyond accuracy are crucial for developing cost-effective techniques. We present the Economical Prompting Index (EPI), a novel metric that combines accuracy scores with token consumption, adjusted by a user-specified cost concern level to reflect different resource constraints. Our study examines 6 advanced prompting techniques, including Chain-of-Thought, Self-Consistency, and Tree of Thoughts, across 10 widely-used language models and 4 diverse datasets. We demonstrate that approaches such as Self-Consistency often provide statistically insignificant gains while becoming cost-prohibitive. For example, on high-performing models like Claude 3.5 Sonnet, the EPI of simpler techniques like Chain-of-Thought (0.72) surpasses more complex methods like Self-Consistency (0.64) at slight cost concern levels. Our findings suggest a reevaluation of complex prompting strategies in resource-constrained scenarios, potentially reshaping future research priorities and improving cost-effectiveness for end-users., Comment: 5 pages (excluding references), accepted to Coling 2025

We report the results of an extensive set of simulations exploring the sensitivity of the BlackCAT CubeSat to long-duration gamma-ray bursts (GRBs). BlackCAT is a NASA APRA-funded CubeSat mission for the detection and real-time sub-arcminute localization of high-redshift ($z\gtrsim 3.5$) GRBs. Thanks to their luminous and long-lived afterglow emissions, GRBs are uniquely valuable probes of high-redshift star-forming galaxies and the intergalactic medium. In addition, each detected GRB with a known redshift serves to localize a region of high-redshift star formation in three dimensions, enabling deep follow-on searches for host galaxies and associated local and large-scale structures. We explore two distinct models for the GRB redshift distribution and luminosity function, both consistent with Swift observations. We find that, for either model, BlackCAT is expected to detect a mean of 42 bursts per year on-orbit, with 6.7% to 10% of these at $z>3.5$. BlackCAT bursts will be localized to $r_{90} \lesssim 55^{\prime\prime}$ precision and reported to the community within seconds. Due to the mission orbit and pointing scheme, bursts will be located in the night sky and well-placed for deep multiwavelength follow-up observations. BlackCAT is on schedule to achieve launch readiness in 2025., Comment: 10 pages, 4 figures, 3 tables; Published in ApJ

Future x-ray observatories will require imaging detectors with fast readout speeds that simultaneously achieve or exceed the other high performance parameters of x-ray charge-coupled devices (CCDs) used in many missions over the past three decades. Fast readout will reduce the impact of pile-up in missions with large collecting areas while also improving performance in other respects like timing resolution. Event-driven readout, in which only pixels with charge from x-ray events are read out, can be used to achieve these faster operating speeds. Speedster-EXD550 detectors are hybrid complementary metal-oxide semiconductor (CMOS) detectors capable of event-driven readout, developed by Teledyne Imaging Sensors and Penn State University. We present initial results from measurements of the first of these detectors, demonstrating their capabilities and performance in both full-frame and event-driven readout modes. These include dark current, read noise, gain variation, and energy resolution measurements from the first two engineering-grade devices., Comment: 14 pages, 7 figures, 4 tables; Published in the Journal of Astronomical Telescopes, Instruments, and Systems

Traditional Statistical Process Control methodologies face several challenges when monitoring defects in complex geometries, such as those of products obtained via Additive Manufacturing techniques. Many approaches cannot be applied in these settings due to the high dimensionality of the data and the lack of parametric and distributional assumptions on the object shapes. Motivated by a case study involving the monitoring of egg-shaped trabecular structures, we investigate two recently-proposed methodologies to detect deviations from the nominal IC model caused by excess or lack of material. Our study focuses on the detection of both isolated large changes in the geometric structure, as well as persistent small deviations. We compare the approach of Scimone et al. (2022) with Zhao and del Castillo (2021) for monitoring defects in a small Phase I sample of 3D-printed objects. While the former control chart is able to detect large defects, the latter allows the detection of nonconforming objects with persistent small defects. Furthermore, we address the fundamental issue of selecting the number of eigenvalues to be monitored in Zhao and del Castillo's method by proposing a dimensionality reduction technique based on kernel principal components. This approach is shown to provide a good detection capability even when considering a large number of eigenvalues. By leveraging the sensitivity of the two monitoring schemes to different magnitudes of nonconformities, we also propose a novel joint monitoring scheme that is capable of identifying both types of defects in the considered case study. Computer code in R and Matlab that implements these methods and replicates the results is available as part of the supplementary material., Comment: We identified some issues in applying the LB method to the data in the study, which impacted the accuracy and reliability of the results. We sincerely thank Professor Enrique del Castillo and Yulin An for their valuable comments and insights, which helped identifying these issues. We appreciate their contributions and are currently in the process of revising and improving the manuscript

Next-generation X-ray observatories, such as the Lynx X-ray Observatory Mission Concept or other similar concepts in the coming decade, will require detectors with high quantum efficiency (QE) across the soft X-ray band to observe the faint objects that drive their mission science objectives. Hybrid CMOS Detectors (HCDs), a form of active-pixel sensor, are promising candidates for use on these missions because of their fast read-out, low power consumption, and intrinsic radiation hardness. In this work, we present QE measurements of a Teledyne H2RG HCD, performed using a gas-flow proportional counter as a reference detector. We find that this detector achieves high QE across the soft X-ray band, with an effective QE of $94.6 \pm 1.1 \%$ at the Mn K$\alpha$/K$\beta$ energies (5.90/6.49 keV), $98.3 \pm 1.9 \%$ at the Al K$\alpha$ energy (1.49 keV), $85.6 \pm 2.8 \%$ at the O K$\alpha$ energy (0.52 keV), and $61.3 \pm 1.1 \%$ at the C K$\alpha$ energy (0.28 keV). These values are in good agreement with our model, based on the absorption of detector layers. We find similar results in a more restrictive analysis considering only high-quality events, with only somewhat reduced QE at lower energies., Comment: 21 pages, 14 figures. Published in the Journal of Astronomical Telescopes, Instruments, and Systems. This work is an expansion of arXiv:2110.01531

The development of data acquisition systems is facilitating the collection of data that are apt to be modelled as functional data. In some applications, the interest lies in the identification of significant differences in group functional means defined by varying experimental conditions, which is known as functional analysis of variance (FANOVA). With real data, it is common that the sample under study is contaminated by some outliers, which can strongly bias the analysis. In this paper, we propose a new robust nonparametric functional ANOVA method (RoFANOVA) that reduces the weights of outlying functional data on the results of the analysis. It is implemented through a permutation test based on a test statistic obtained via a functional extension of the classical robust $ M $-estimator. By means of an extensive Monte Carlo simulation study, the proposed test is compared with some alternatives already presented in the literature, in both one-way and two-way designs. The performance of the RoFANOVA is demonstrated in the framework of a motivating real-case study in the field of additive manufacturing that deals with the analysis of spatter ejections. The RoFANOVA method is implemented in the R package rofanova, available online at https://github.com/unina-sfere/rofanova.

Next-generation X-ray observatories, such as the Lynx X-ray Observatory Mission Concept, will require detectors with high quantum efficiency (QE) across the soft X-ray band to observe the faint objects that drive their mission science cases. Hybrid CMOS Detectors (HCDs), a form of active-pixel sensor, are promising candidates for use on these missions because of their faster read-out, lower power consumption, and greater radiation hardness than detectors used in the current generation of X-ray telescopes. In this work, we present QE measurements of a Teledyne H2RG HCD. These measurements were performed using a gas-flow proportional counter as a reference detector to measure the absolute flux incident on the HCD. We find an effective QE of $95.0 \pm 1.1\%$ at the Mn K$\alpha$/K$\beta$ lines (at 5.9 and 6.5 keV), $98.5 \pm 1.8\%$ at the Al K$\alpha$ line (1.5 keV), and $85.0 \pm 2.8\%$ at the O K$\alpha$ line (0.52 keV)., Comment: 11 pages, 8 figures, appears in Proc. SPIE 2021

We conduct X-ray spectral fits on 184 likely counterparts to Fermi-LAT 3FGL unassociated sources. Characterization and classification of these sources allows for more complete population studies of the high-energy sky. Most of these X-ray spectra are well fit by an absorbed power law model, as expected for a population dominated by blazars and pulsars. A small subset of 7 X-ray sources have spectra unlike the power law expected from a blazar or pulsar and may be linked to coincident stars or background emission. We develop a multiwavelength machine learning classifier to categorize unassociated sources into pulsars and blazars using gamma- and X-ray observations. Training a random forest procedure with known pulsars and blazars, we achieve a cross-validated classification accuracy of 98.6%. Applying the random forest routine to the unassociated sources returned 126 likely blazar candidates (defined as $ P_{bzr} > 90 \% $) and 5 likely pulsar candidates ($ P_{bzr} < 10 \% $). Our new X-ray spectral analysis does not drastically alter the random forest classifications of these sources compared to previous works, but it builds a more robust classification scheme and highlights the importance of X-ray spectral fitting. Our procedure can be further expanded with UV, visual, or radio spectral parameters or by measuring flux variability., Comment: 11 pages text, 4 figures, 4 tables (2 in-text, 2 in 11-page appendix), accepted for publication in the Astronomical Journal

The use of video-imaging data for in-line process monitoring applications has become more and more popular in the industry. In this framework, spatio-temporal statistical process monitoring methods are needed to capture the relevant information content and signal possible out-of-control states. Video-imaging data are characterized by a spatio-temporal variability structure that depends on the underlying phenomenon, and typical out-of-control patterns are related to the events that are localized both in time and space. In this paper, we propose an integrated spatio-temporal decomposition and regression approach for anomaly detection in video-imaging data. Out-of-control events are typically sparse spatially clustered and temporally consistent. Therefore, the goal is to not only detect the anomaly as quickly as possible ("when") but also locate it ("where"). The proposed approach works by decomposing the original spatio-temporal data into random natural events, sparse spatially clustered and temporally consistent anomalous events, and random noise. Recursive estimation procedures for spatio-temporal regression are presented to enable the real-time implementation of the proposed methodology. Finally, a likelihood ratio test procedure is proposed to detect when and where the hotspot happens. The proposed approach was applied to the analysis of video-imaging data to detect and locate local over-heating phenomena ("hotspots") during the layer-wise process in a metal additive manufacturing process.

Usually, Neural Networks models are trained with a large dataset of images in homogeneous backgrounds. The issue is that the performance of the network models trained could be significantly degraded in a complex and heterogeneous environment. To mitigate the issue, this paper develops a framework that permits to autonomously generate a training dataset in heterogeneous cluttered backgrounds. It is clear that the learning effectiveness of the proposed framework should be improved in complex and heterogeneous environments, compared with the ones with the typical dataset. In our framework, a state-of-the-art image segmentation technique called DeepLab is used to extract objects of interest from a picture and Chroma-key technique is then used to merge the extracted objects of interest into specific heterogeneous backgrounds. The performance of the proposed framework is investigated through empirical tests and compared with that of the model trained with the COCO dataset. The results show that the proposed framework outperforms the model compared. This implies that the learning effectiveness of the framework developed is superior to the models with the typical dataset., Comment: IEEE RED-UAS 2019 Conference