Search

This study analyzed Pennsylvania School Climate Survey data from students and staff in the 2021/22 school year to assess the validity and reliability of the elementary school student version of the survey; approaches to scoring the survey in individual schools at all grade levels; and perceptions of school climate across student, staff, and school groups. The survey encourages data-informed efforts in participating Pennsylvania schools to foster supportive learning environments that promote social and emotional wellness for students and staff. The study validated the elementary school student survey but found that one domain--safe and respectful school climate--did not meet the reliability threshold and thus suggests that revisions are needed. At all grade levels noninstructional staff had the most positive perceptions of school climate, followed by classroom teachers then students. The study found that different approaches to combining the school climate scores of students, teachers, and noninstructional staff within schools yielded slightly different distributions of school climate summary index scores. It also found that different performance category thresholds resulted in similar distributions of schools across categories. Scores calculated using simple averages were strongly and positively correlated with scores calculated using a more complex approach (Rasch models), suggesting that both approaches deliver similar information. School climate scores varied across student groups (defined by race/ethnicity, gender, and grade level) within schools and across school groups. Larger schools and schools with higher percentages of Black students tended to have lower school climate scores than other schools. The findings can inform the Pennsylvania Department of Education's decisionmaking on revisions to the elementary school student survey, approaches to scoring and reporting survey results, and efforts to increase participation in future survey administrations.

We propose a scaling approach to evaluating the qubit resources required by concatenated fault-tolerant quantum computing. The approach gives closed-form expressions, which remain simple for multiple levels of concatenation, making it an ideal tool to compare and minimize the resource costs of different concatenation schemes. We then use it to study the resources required for normal and magic operations. Here, magic operations require the preparation, verification and injection of complicated states called "magic states", while normal operations do not. It is often expected that magic operations will dominate a computation's physical resource requirements (qubits, gates, etc.), although this expectation has been cast in doubt for surface codes. Our results show that this expectation is also wrong for typical concatenated codes, with magic operations rarely being costly. We give concrete examples for the concatenated 7-qubit scheme with Steane error-correction gadgets or a flag-qubits approach. While optimizations that affect all operations are naturally more effective than ones that affect only magic operations, the surprising conclusion of our work is that the former can reduce costs by several orders of magnitude whereas the latter contributes only marginal reductions. This is particularly surprising given the numerous works on optimizations that affect only magic operations., Comment: 43 pages (starting with a 5 page summary of results), 19 figures

Understanding the nature of dark matter in the Universe is an important goal of modern cosmology. A key method for probing this distribution is via weak gravitational lensing mass-mapping - a challenging ill-posed inverse problem where one infers the convergence field from observed shear measurements. Upcoming stage IV surveys, such as those made by the Vera C. Rubin Observatory and Euclid satellite, will provide a greater quantity and precision of data for lensing analyses, necessitating high-fidelity mass-mapping methods that are computationally efficient and that also provide uncertainties for integration into downstream cosmological analyses. In this work we introduce MMGAN, a novel mass-mapping method based on a regularised conditional generative adversarial network (GAN) framework, which generates approximate posterior samples of the convergence field given shear data. We adopt Wasserstein GANs to improve training stability and apply regularisation techniques to overcome mode collapse, issues that otherwise are particularly acute for conditional GANs. We train and validate our model on a mock COSMOS-style dataset before applying it to true COSMOS survey data. Our approach significantly outperforms the Kaiser-Squires technique and achieves similar reconstruction fidelity as alternative state-of-the-art deep learning approaches. Notably, while alternative approaches for generating samples from a learned posterior are slow (e.g. requiring $\sim$10 GPU minutes per posterior sample), MMGAN can produce a high-quality convergence sample in less than a second.

The magnetised near detector (ND280) of the T2K long-baseline neutrino oscillation experiment has been recently upgraded aiming to satisfy the requirement of reducing the systematic uncertainty from measuring the neutrinonucleus interaction cross section, which is the largest systematic uncertainty in the search for leptonic charge-parity symmetry violation. A key component of the upgrade is SuperFGD, a 3D segmented plastic scintillator detector made of approximately 2,000,000 optically-isolated 1 cm3 cubes. It will provide a 3D image of GeV neutrino interactions by combining tracking and stopping power measurements of final state particles with sub-nanosecond time resolution. The performance of SuperFGD is characterized by the precision of its response to charged particles as well as the systematic effects that might affect the physics measurements. Hence, a detailed Geant4 based optical simulation of the SuperFGD building block, i.e. a plastic scintillating cube read out by three wavelength shifting fibers, has been developed and validated with the different datasets collected in various beam tests. In this manuscript the description of the optical model as well as the comparison with data are reported., Comment: 31 pages, 15 figures

Interrupted X-ray computed tomography (X-CT) has been the common way to observe the deformation of materials during an experiment. While this approach is effective for quasi-static experiments, it has never been possible to reconstruct a full 3d tomography during a dynamic experiment which cannot be interrupted. In this work, we propose that neural rendering tools can be used to drive the paradigm shift to enable 3d reconstruction during dynamic events. First, we derive theoretical results to support the selection of projections angles. Via a combination of synthetic and experimental data, we demonstrate that neural radiance fields can reconstruct data modalities of interest more efficiently than conventional reconstruction methods. Finally, we develop a spatio-temporal model with spline-based deformation field and demonstrate that such model can reconstruct the spatio-temporal deformation of lattice samples in real-world experiments., Comment: 24 pages, 14 figures. Submitted to NeurIPS 2024 ML4PS. For associated visualizations, see https://neural-xray.github.io/nerfxray

GPT-4o is an autoregressive omni model that accepts as input any combination of text, audio, image, and video, and generates any combination of text, audio, and image outputs. It's trained end-to-end across text, vision, and audio, meaning all inputs and outputs are processed by the same neural network. GPT-4o can respond to audio inputs in as little as 232 milliseconds, with an average of 320 milliseconds, which is similar to human response time in conversation. It matches GPT-4 Turbo performance on text in English and code, with significant improvement on text in non-English languages, while also being much faster and 50\% cheaper in the API. GPT-4o is especially better at vision and audio understanding compared to existing models. In line with our commitment to building AI safely and consistent with our voluntary commitments to the White House, we are sharing the GPT-4o System Card, which includes our Preparedness Framework evaluations. In this System Card, we provide a detailed look at GPT-4o's capabilities, limitations, and safety evaluations across multiple categories, focusing on speech-to-speech while also evaluating text and image capabilities, and measures we've implemented to ensure the model is safe and aligned. We also include third-party assessments on dangerous capabilities, as well as discussion of potential societal impacts of GPT-4o's text and vision capabilities.

When training transformers on graph-structured data, incorporating information about the underlying topology is crucial for good performance. Topological masking, a type of relative position encoding, achieves this by upweighting or downweighting attention depending on the relationship between the query and keys in a graph. In this paper, we propose to parameterise topological masks as a learnable function of a weighted adjacency matrix -- a novel, flexible approach which incorporates a strong structural inductive bias. By approximating this mask with graph random features (for which we prove the first known concentration bounds), we show how this can be made fully compatible with linear attention, preserving $\mathcal{O}(N)$ time and space complexity with respect to the number of input tokens. The fastest previous alternative was $\mathcal{O}(N \log N)$ and only suitable for specific graphs. Our efficient masking algorithms provide strong performance gains for tasks on image and point cloud data, including with $>30$k nodes.

We provide a non-asymptotic analysis of the linear instrumental variable estimator allowing for the presence of exogeneous covariates. In addition, we introduce a novel measure of the strength of an instrument that can be used to derive non-asymptotic confidence intervals. For strong instruments, these non-asymptotic intervals match the asymptotic ones exactly up to higher order corrections; for weaker instruments, our intervals involve adaptive adjustments to the instrument strength, and thus remain valid even when asymptotic predictions break down. We illustrate our results via an analysis of the effect of PM2.5 pollution on various health conditions, using wildfire smoke exposure as an instrument. Our analysis shows that exposure to PM2.5 pollution leads to statistically significant increases in incidence of health conditions such as asthma, heart disease, and strokes.

Generative Artificial Intelligence (GenAI) is becoming ubiquitous in our daily lives. The increase in computational power and data availability has led to a proliferation of both single- and multi-modal models. As the GenAI ecosystem matures, the need for extensible and model-agnostic risk identification frameworks is growing. To meet this need, we introduce the Python Risk Identification Toolkit (PyRIT), an open-source framework designed to enhance red teaming efforts in GenAI systems. PyRIT is a model- and platform-agnostic tool that enables red teamers to probe for and identify novel harms, risks, and jailbreaks in multimodal generative AI models. Its composable architecture facilitates the reuse of core building blocks and allows for extensibility to future models and modalities. This paper details the challenges specific to red teaming generative AI systems, the development and features of PyRIT, and its practical applications in real-world scenarios.

Nucleon structure functions, as measured in lepton-nucleon scattering, have historically provided a critical observable in the study of partonic dynamics within the nucleon. However, at very large parton momenta it is both experimentally and theoretically challenging to extract parton distributions due to the probable onset of non-perturbative contributions and the unavailability of high precision data at critical kinematics. Extraction of the neutron structure and the d-quark distribution have been further challenging due to the necessity of applying nuclear corrections when utilizing scattering data from a deuteron target to extract free neutron structure. However, a program of experiments has been carried out recently at the energy-upgraded Jefferson Lab electron accelerator aimed at significantly reducing the nuclear correction uncertainties on the d-quark distribution function at large partonic momentum. This allows leveraging the vast body of deuterium data covering a large kinematic range to be utilized for d-quark parton distribution function extraction. We present new data from experiment E12-10-002 carried out in Jefferson Lab Hall C on the deuteron to proton cross-section ratio at large BJorken-x. These results significantly improve the precision of existing data, and provide a first look at the expected impact on quark distributions extracted from global parton distribution function fits.

Anonymity is an important principle online. However, malicious actors have long used misleading identities to conduct fraud, spread disinformation, and carry out other deceptive schemes. With the advent of increasingly capable AI, bad actors can amplify the potential scale and effectiveness of their operations, intensifying the challenge of balancing anonymity and trustworthiness online. In this paper, we analyze the value of a new tool to address this challenge: "personhood credentials" (PHCs), digital credentials that empower users to demonstrate that they are real people -- not AIs -- to online services, without disclosing any personal information. Such credentials can be issued by a range of trusted institutions -- governments or otherwise. A PHC system, according to our definition, could be local or global, and does not need to be biometrics-based. Two trends in AI contribute to the urgency of the challenge: AI's increasing indistinguishability from people online (i.e., lifelike content and avatars, agentic activity), and AI's increasing scalability (i.e., cost-effectiveness, accessibility). Drawing on a long history of research into anonymous credentials and "proof-of-personhood" systems, personhood credentials give people a way to signal their trustworthiness on online platforms, and offer service providers new tools for reducing misuse by bad actors. In contrast, existing countermeasures to automated deception -- such as CAPTCHAs -- are inadequate against sophisticated AI, while stringent identity verification solutions are insufficiently private for many use-cases. After surveying the benefits of personhood credentials, we also examine deployment risks and design challenges. We conclude with actionable next steps for policymakers, technologists, and standards bodies to consider in consultation with the public., Comment: 63 pages, 7 figures, 5 tables; minor additions to acknowledgments and wording changes for clarity; corrected typo

The goal of this paper is to study the Rees algebra $\mathfrak{R}(I)$and the special fiber ring $\mathfrak{F}(I)$ for a family of ideals. Let $R=\mathbb{K}[x_1, \ldots, x_d]$ with $d\geq 4$ be a polynomial ring with homogeneous maximal ideal $\mathfrak{m}$. We study the $R$-ideals $I$, which are $\mathfrak{m}$-primary, Gorenstein, generated in degree 2, and have a Gorenstein linear resolution. In the smallest case, $d=4$, this family includes the ideals of $2\times 2$ minors of a general $3\times 3$ matrix of linear forms in $R$. We show that the defining ideal of the Rees algebra will be of fiber type. That is, the defining ideal of the Rees algebra is generated by the defining ideals of the special fiber ring and of the symmetric algebra. We use the fact that these ideals differ from $\mathfrak{m}^2$ by exactly one minimal generator to describe the defining ideal $\mathfrak{F}(I)$ as a sub-ideal of the defining ideal of $\mathfrak{F}(\mathfrak{m}^2)$, which is well known to be the ideal of $2\times 2$ minors of a symmetric matrix of variables.

Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. Our largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. This paper presents an extensive empirical evaluation of Llama 3. We find that Llama 3 delivers comparable quality to leading language models such as GPT-4 on a plethora of tasks. We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. We observe this approach performs competitively with the state-of-the-art on image, video, and speech recognition tasks. The resulting models are not yet being broadly released as they are still under development.

Spatially-resolved heating and cooling in nanostructures is nowadays measured with various nanoscale thermometry techniques, including scanning thermometry. Yet the most commonly used theory of nanoscale heating and thermoelectricity -- Landauer scattering theory -- is not appropriate to model such measurements. Hence, we analyze a minimal model of spatially-resolved heat transfer between electrons and phonons in simple thermoelectric nanostructures. This combines Landauer scattering formalism with a Boltzmann equation for transport, revealing the non-locality of Joule heating and Peltier cooling induced by a scatterer in a nanowire. The corresponding heating or cooling of the phonons is caused by the voltage drop at the scatterer, but is often maximal at a certain distance from the scatterer. This distance is of the order of the electron-phonon scattering length. Scanning thermal microscopy, such as SQUID-on-tip thermometers, should detect this non-locality as phonon hot spots and cold spots, spatially separated from the scatterer. We provide physical arguments explaining the thermoelectric response of the combined system of wire and scatterer, and in particular, why the resulting heating and cooling is sometimes the opposite to that predicted by the standard Landauer scattering theory., Comment: Version 2: added 2 references, some other very minor improvements

We present High-Density Visual Particle Dynamics (HD-VPD), a learned world model that can emulate the physical dynamics of real scenes by processing massive latent point clouds containing 100K+ particles. To enable efficiency at this scale, we introduce a novel family of Point Cloud Transformers (PCTs) called Interlacers leveraging intertwined linear-attention Performer layers and graph-based neighbour attention layers. We demonstrate the capabilities of HD-VPD by modeling the dynamics of high degree-of-freedom bi-manual robots with two RGB-D cameras. Compared to the previous graph neural network approach, our Interlacer dynamics is twice as fast with the same prediction quality, and can achieve higher quality using 4x as many particles. We illustrate how HD-VPD can evaluate motion plan quality with robotic box pushing and can grasping tasks. See videos and particle dynamics rendered by HD-VPD at https://sites.google.com/view/hd-vpd.

Marine debris poses a significant ecological threat to birds, fish, and other animal life. Traditional methods for assessing debris accumulation involve labor-intensive and costly manual surveys. This study introduces a framework that utilizes aerial imagery captured by drones to conduct remote trash surveys. Leveraging computer vision techniques, our approach detects, classifies, and maps marine debris distributions. The framework uses Grounding DINO, a transformer-based zero-shot object detector, and CLIP, a vision-language model for zero-shot object classification, enabling the detection and classification of debris objects based on material type without the need for training labels. To mitigate over-counting due to different views of the same object, Scale-Invariant Feature Transform (SIFT) is employed for duplicate matching using local object features. Additionally, we have developed a user-friendly web application that facilitates end-to-end analysis of drone images, including object detection, classification, and visualization on a map to support cleanup efforts. Our method achieves competitive performance in detection (0.69 mean IoU) and classification (0.74 F1 score) across seven debris object classes without labeled data, comparable to state-of-the-art supervised methods. This framework has the potential to streamline automated trash sampling surveys, fostering efficient and sustainable community-led cleanup initiatives., Comment: 9 pages, 6 figures, 2 tables, In Proceedings of the 7th ACM SIGCAS/SIGCHI Conference on Computing and Sustainable Societies

Simulating large scenes with many rigid objects is crucial for a variety of applications, such as robotics, engineering, film and video games. Rigid interactions are notoriously hard to model: small changes to the initial state or the simulation parameters can lead to large changes in the final state. Recently, learned simulators based on graph networks (GNNs) were developed as an alternative to hand-designed simulators like MuJoCo and PyBullet. They are able to accurately capture dynamics of real objects directly from real-world observations. However, current state-of-the-art learned simulators operate on meshes and scale poorly to scenes with many objects or detailed shapes. Here we present SDF-Sim, the first learned rigid-body simulator designed for scale. We use learned signed-distance functions (SDFs) to represent the object shapes and to speed up distance computation. We design the simulator to leverage SDFs and avoid the fundamental bottleneck of the previous simulators associated with collision detection. For the first time in literature, we demonstrate that we can scale the GNN-based simulators to scenes with hundreds of objects and up to 1.1 million nodes, where mesh-based approaches run out of memory. Finally, we show that SDF-Sim can be applied to real world scenes by extracting SDFs from multi-view images.

Understanding the large-scale structure of the Universe and unravelling the mysteries of dark matter are fundamental challenges in contemporary cosmology. Reconstruction of the cosmological matter distribution from lensing observables, referred to as 'mass-mapping' is an important aspect of this quest. Mass-mapping is an ill-posed problem, meaning there is inherent uncertainty in any convergence map reconstruction. The demand for fast and efficient reconstruction techniques is rising as we prepare for upcoming surveys. We present a novel approach which utilises deep learning, in particular a conditional Generative Adversarial Network (cGAN), to approximate samples from a Bayesian posterior distribution, meaning they can be interpreted in a statistically robust manner. By combining data-driven priors with recent regularisation techniques, we introduce an approach that facilitates the swift generation of high-fidelity, mass maps. Furthermore, to validate the effectiveness of our approach, we train the model on mock COSMOS-style data, generated using Colombia Lensing's kappaTNG mock weak lensing suite. These preliminary results showcase compelling convergence map reconstructions and ongoing refinement efforts are underway to enhance the robustness of our method further., Comment: 2 pages, 1 figure, submitted for conference proceedings for '58th Recontres de Moriond', 2024

Modern diffusion MRI sequences commonly acquire a large number of volumes with diffusion sensitization gradients of differing strengths or directions. Such sequences rely on echo-planar imaging (EPI) to achieve reasonable scan duration. However, EPI is vulnerable to off-resonance effects, leading to tissue susceptibility and eddy-current induced distortions. The latter is particularly problematic because it causes misalignment between volumes, disrupting downstream modelling and analysis. The essential correction of eddy distortions is typically done post-acquisition, with image registration. However, this is non-trivial because correspondence between volumes can be severely disrupted due to volume-specific signal attenuations induced by varying directions and strengths of the applied gradients. This challenge has been successfully addressed by the popular FSL~Eddy tool but at considerable computational cost. We propose an alternative approach, leveraging recent advances in image processing enabled by deep learning (DL). It consists of two convolutional neural networks: 1) An image translator to restore correspondence between images; 2) A registration model to align the translated images. Results demonstrate comparable distortion estimates to FSL~Eddy, while requiring only modest training sample sizes. This work, to the best of our knowledge, is the first to tackle this problem with deep learning. Together with recently developed DL-based susceptibility correction techniques, they pave the way for real-time preprocessing of diffusion MRI, facilitating its wider uptake in the clinic., Comment: accepted in MICCAI 2024 conference

Value-at-risk (VaR) and expected shortfall (ES) are two commonly utilized metrics for quantifying financial risk. In this study, we review the widely employed Generalized Autoregressive Conditional Heteroskedasticity (GARCH) models. These models are explored with diverse distributional assumptions on innovation, including parametric, non-parametric, and `semi-parametric' that incorporates a parametric tail distribution based on extreme value theory. Additionally, we introduce a non-parametric local linear quantile autoregression (LLQAR) with kernel weights depending on the distance between the current loss and past losses, and decreasing in the time lag. To evaluate the performance of different methods for VaR and ES estimation, we employ a multi-criteria approach. This involves mean squared error assessment using simulated data, backtesting on both simulated data and US stocks, and application of the ESBootstrap test. The LLQAR method, which does not necessarily require stationarity assumptions, seems to perform better for simulated non-stationary data as well as real-world data, for estimating VaR and ES., Comment: 54 pages, 21 figures

Machine learning systems require representations of the real world for training and testing - they require data, and lots of it. Collecting data at scale has logistical and ethical challenges, and synthetic data promises a solution to these challenges. Instead of needing to collect photos of real people's faces to train a facial recognition system, a model creator could create and use photo-realistic, synthetic faces. The comparative ease of generating this synthetic data rather than relying on collecting data has made it a common practice. We present two key risks of using synthetic data in model development. First, we detail the high risk of false confidence when using synthetic data to increase dataset diversity and representation. We base this in the examination of a real world use-case of synthetic data, where synthetic datasets were generated for an evaluation of facial recognition technology. Second, we examine how using synthetic data risks circumventing consent for data usage. We illustrate this by considering the importance of consent to the U.S. Federal Trade Commission's regulation of data collection and affected models. Finally, we discuss how these two risks exemplify how synthetic data complicates existing governance and ethical practice; by decoupling data from those it impacts, synthetic data is prone to consolidating power away those most impacted by algorithmically-mediated harm.

With the support of the national program on measurements, standards, and evaluation of quantum technologies MetriQs-France, a part of the French national quantum strategy, the BACQ project is dedicated to application-oriented benchmarks for quantum computing. The consortium gathering THALES, EVIDEN, an Atos business, CEA, CNRS, TERATEC, and LNE aims at establishing performance evaluation criteria of reference, meaningful for industry users.

The Pennsylvania Department of Education's (PDE's) Office for Safe Schools partnered with REL Mid-Atlantic to conduct a study using data from PDE's school climate survey. This survey, which is available on a voluntary basis to any school in the state, provides a way to track school climate and identify schools that need additional support to improve school climate. The REL study analyzed changes in scores from a pre-pandemic year (2018/19) to the 2020/21 and 2021/22 school years, both of which were affected by COVID-19 and related policy changes. In a sample of Pennsylvania public schools that took the survey consistently across multiple years, students and teachers reported more positive perceptions of school climate in the 2020/21 school year, during hybrid and remote learning, compared to 2018/19 and 2021/22. This was an unexpected positive bump in the year in which schools experienced the most pandemic-related disruption. In contrast, school climate scores were steady across the years before COVID-19. The study also found no evidence of a significant decline in school climate scores between 2018/19 and 2021/22, suggesting the pandemic did not have a lasting negative effect on school climate in this sample of schools. One important caveat of this study is that the sample of schools was small and not representative of the rest of the state of Pennsylvania. The study also examined the relationship between school climate scores in 2021/22 and the amount of virtual and hybrid instruction used during 2020/21 and found no association. However, this analysis was limited by the small sample size and could only rule out a large association; a small or moderate association is still possible. [For the appendixes, see ED629791.]

These are the appendixes for the full report, "Changes in School Climate during COVID-19 in a Sample of Pennsylvania Schools." The REL study analyzed changes in scores from a pre-pandemic year (2018/19) to the 2020/21 and 2021/22 school years, both of which were affected by COVID-19 and related policy changes. These technical appendixes are meant to supplement the slideshow to provide extended detail on the methods and analyses used to answer the study's research questions. Appendixes: (1) Data and methods; (2) Supporting analysis; and (3) Nonresponsive bias analysis. [For the full report, see ED629789.]