Search

Machine learning based surrogate models offer researchers powerful tools for accelerating simulation-based workflows. However, as standard datasets in this space often cover small classes of physical behavior, it can be difficult to evaluate the efficacy of new approaches. To address this gap, we introduce the Well: a large-scale collection of datasets containing numerical simulations of a wide variety of spatiotemporal physical systems. The Well draws from domain experts and numerical software developers to provide 15TB of data across 16 datasets covering diverse domains such as biological systems, fluid dynamics, acoustic scattering, as well as magneto-hydrodynamic simulations of extra-galactic fluids or supernova explosions. These datasets can be used individually or as part of a broader benchmark suite. To facilitate usage of the Well, we provide a unified PyTorch interface for training and evaluating models. We demonstrate the function of this library by introducing example baselines that highlight the new challenges posed by the complex dynamics of the Well. The code and data is available at https://github.com/PolymathicAI/the_well., Comment: 38th Conference on Neural Information Processing Systems (NeurIPS 2024) Track on Datasets and Benchmarks

Each year, the production of synthetic dye wastewater reaches a trillion tons, posing a significant challenge to addressing water scarcity on a global level. Hence, the treatment of wastewater to prevent water scarcity is of prime importance, and failing to do so will increase ecotoxicological risks and human health. Textile wastewater contains harmful dye. Photocatalytic degradation of such dye-contaminated wastewater is crucial to purifying the dye-contaminated water. However, this process takes time, uses high-power lamps, and is expensive. Here, we report the effect of the concentration of precursor on the size and surface morphology of TiO2 nanostructures prepared by facile hydrothermal synthesis and its ability to perform as a photocatalyst to degrade the most common industrial textile dye, methylene blue (MB), under natural sunlight. The impact of particle size on the photocatalytic activity and photocarrier migration rate was thoroughly examined. Also, the effect of pH on adsorption and photocatalytic degradation has been evaluated in detail. With several optimized conditions, almost complete dye degradation was achieved within 40 minutes under the direct illumination of natural sunlight. The enhanced photocatalytic performance can be correlated to the synergetic effect of a higher charge transfer mechanism, good catalytic active surface area availability (386 m2/g), and several optimized parameters that affect the reaction efficacy. Additionally, repeated use of NPs without sacrificing performance five times confirmed its stability and Sustainability as a promising candidate for large-scale industrial textile wastewater remedies.

We analyze the timing of NGO campaigns to shed light on NGOs' objectives and how their strategies evolve over time. Using data from 2,500 campaigns, we find that NGOs are six times more likely to launch campaigns on their target's Annual General Meeting (AGM) date. Although this strategy increases media exposure and stakeholder scrutiny, resulting in consumer boycotts and related shareholder proposals at the following AGM of the targeted firm, it has no impact on current AGM votes. As NGOs build reputational capital, they adjust their timing to influence AGM votes, revealing the trade-offs they face to drive corporate change.

In this study, we present the integration of a neural network-based parameterization into the global atmospheric model ARP-GEM1, leveraging the Python interface of the OASIS coupler. This approach facilitates the exchange of fields between the Fortran-based ARP-GEM1 model and a Python component responsible for neural network inference. As a proof-of-concept experiment, we trained a neural network to emulate the deep convection parameterization of ARP-GEM1. Using the flexible Fortran/Python interface, we have successfully replaced ARP-GEM1's deep convection scheme with a neural network emulator. To assess the performance of the neural network deep convection scheme, we have run a 5-years ARP-GEM1 simulation using the neural network emulator. The evaluation of averaged fields showed good agreement with output from an ARP-GEM1 simulation using the physics-based deep convection scheme. The Python component was deployed on a separate partition from the general circulation model, using GPUs to increase inference speed of the neural network., Comment: 10 pages, 5 figures, submitted to Artificial Intelligence for the Earth Systems

We provide a unified framework in which the three emotions at the heart of narrative tension (curiosity, suspense and surprise) are formalized. This framework is built on nonmonotonic reasoning which allows us to compactly represent the default behavior of the world and to simulate the affective evolution of an agent receiving a story. After formalizing the notions of awareness, curiosity, surprise and suspense, we explore the properties induced by our definitions and study the computational complexity of detecting them. We finally propose means to evaluate these emotions' intensity for a given agent listening to a story.

In this paper, the introduction of interference cancellation in full-duplex joint radar and communication receivers is analysed. More specifically, a focus is made on scenarios in which the receiver simultaneously receives radar echoes from the environment, and communication signals from other joint radar and communication transceivers. First, the phase-coded frequency modulated continuous wave waveform designed for integrated sensing and communication is presented. Then, simple structures in which interference cancellation is only applied at the radar or communication function are proposed, relying on the information gathered at the other function. The detection probability and bit error rate improvements are analysed numerically w.r.t. different system parameters, such as the communication constellation, or the number of transmitted pulses. The introduction of error correcting codes is also considered. Next, iterative interference cancellation structures are investigated. Thanks to multiple interference cancellation layers, both the radar and communication performance are improved, and the robustness of the system to any scenario is enhanced. This is shown numerically through the analysis of the performance achieved by simple and iterative structures, which are compared for different radar echo to communication signal power ratio. Finally, a dynamic automotive scenario is considered. Leveraging on previous radar measurements, the parameters of the radar echo are inferred, and it is reconstructed beforehand, enabling to remove the first radar processing block. The complexity of the iterative structures is thus reduced, at the price of a slight performance reduction. A dynamic automotive scenario is considered, highlighting the impact of the tracking of the next vehicle ahead on the system.

This is the second part of a series of two articles focused on the development and evaluation of the ARP-GEM1 global atmosphere model. The first paper introduced the model's new physics and speedup improvements. In this second part, we evaluate ARP-GEM1 through a set of 30-year prescribed sea-surface-temperature simulations at 55, 25, 12.6, and 6.3 km resolutions. The model demonstrates reliability in representing global climate metrics, comparable to the top-performing CMIP6 models, while maintaining high computational efficiency at all resolutions. These simulations further demonstrate the feasibility of O(10)-km climate simulations, and highlight their added value, particularly for capturing phenomena such as cyclones. Ultimately, these exploratory simulations should be considered an intermediate step toward the development and tuning of even higher-resolution, convection-permitting kilometer-scale configurations., Comment: 16 pages, 16 figures, submitted to Journal of Climate

This is the first part of a series of two articles describing the ARP-GEM global atmosphere model version 1 and its evaluation in simulations from 55 km to 6 km resolutions. This article provides a complete description of ARP-GEM1, focusing on its new physical parameterizations and acceleration factors aimed at improving computational efficiency. ARP-GEM1 is approximately 15 times faster than its base model version, ARPEGE-Climat v6.3, while maintaining accurate simulations and enhancing model performance, as shown in Part II.This significant acceleration results from a combination of optimizations, rather than a single factor, with each factor's contribution quantified. Additionally, a detailed decomposition of the model's speedup per component is presented. Streamlining the model reduces computational costs, enhances transparency and portability, and maximizes the benefits of future advanced computing technologies. The results presented here suggest that kilometer-scale global climate simulations should become feasible in the near future., Comment: 20 pages, 4 figures, submitted to Journal of Climate

This study introduces a novel method to investigate in-situ light transport within optically thick ensembles of cold atoms, exploiting the internal structure of alkaline-earth metals. A method for creating an optical excitation at the center of a large atomic cloud is demonstrated, and we observe its propagation through multiple scattering events. In conditions where the cloud size is significantly larger than the transport mean free path, a diffusive regime is identified. We measure key parameters including the diffusion coefficient, transport velocity, and transport time, finding a good agreement with diffusion models. We also demonstrate that the frequency of the photons launched inside the system can be controlled. This approach enables direct time- and space-resolved observation of light diffusion in atomic ensembles, offering a promising avenue for exploring new diffusion regimes.

History eXplanation based on Predicates (HXP), studies the behavior of a Reinforcement Learning (RL) agent in a sequence of agent's interactions with the environment (a history), through the prism of an arbitrary predicate. To this end, an action importance score is computed for each action in the history. The explanation consists in displaying the most important actions to the user. As the calculation of an action's importance is #W[1]-hard, it is necessary for long histories to approximate the scores, at the expense of their quality. We therefore propose a new HXP method, called Backward-HXP, to provide explanations for these histories without having to approximate scores. Experiments show the ability of B-HXP to summarise long histories.

Photovoltaic (PV) energy grows rapidly and is crucial for the decarbonization of electric systems. However, centralized registries recording the technical characteristifs of rooftop PV systems are often missing, making it difficult to accurately monitor this growth. The lack of monitoring could threaten the integration of PV energy into the grid. To avoid this situation, the remote sensing of rooftop PV systems using deep learning emerged as a promising solution. However, existing techniques are not reliable enough to be used by public authorities or transmission system operators (TSOs) to construct up-to-date statistics on the rooftop PV fleet. The lack of reliability comes from the fact that deep learning models are sensitive to distribution shifts. This work proposes a comprehensive evaluation of the effects of distribution shifts on the classification accuracy of deep learning models trained to detect rooftop PV panels on overhead imagery. We construct a benchmark to isolate the sources of distribution shift and introduce a novel methodology that leverages explainable artificial intelligence (XAI) and decomposition of the input image and model's decision in terms of scales to understand how distribution shifts affect deep learning models. Finally, based on our analysis, we introduce a data augmentation technique meant to improve the robustness of deep learning classifiers to varying acquisition conditions. We show that our proposed approach outperforms competing methods. We discuss some practical recommendations for mapping PV systems using overhead imagery and deep learning models., Comment: 24 pages, 13 figures, 5 tables, manuscript submitted to Environmental Data Science

In this note, we show that the sequence of maximum values in frieze patterns of type $A_n$ is the sequence of Fibonacci numbers, and that of frieze patterns of type $C_n$ is the sequence of odd Fibonacci numbers., Comment: Latest version with upper bounds in types An and Cn. Note associated to a Lean4 formalisation project available on GitHub

We present extensive photometric and spectroscopic observations of the nearby Type Ia supernova (SN) 2023wrk at a distance of about 40 Mpc. The earliest detection of this SN can be traced back to a few hours after the explosion. Within the first few days the light curve shows a bump feature, while the B - V color is blue and remains nearly constant. The overall spectral evolution is similar to that of an SN 1991T/SN 1999aa-like SN Ia, while the C II $\lambda6580$ absorption line appears to be unusually strong in the first spectrum taken at $t \approx -$15.4 days after the maximum light. This carbon feature disappears quickly in subsequent evolution but it reappears at around the time of peak brightness. The complex evolution of the carbon line and the possible detection of Ni III absorption around 4700 {\AA} and 5300 {\AA} in the earliest spectra indicate macroscopic mixing of fuel and ash. The strong carbon lines is likely related to collision of SN ejecta with unbound carbon, consistent with the predictions of pulsational delayed-detonation or carbon-rich circumstellar-matter interaction models. Among those carbon-rich SNe Ia with strong C II $\lambda6580$ absorption at very early times, the line-strength ratio of C II to Si II and the B-V color evolution are found to exhibit large diversity, which may be attributed to different properties of unbound carbon and outward-mixing $^{56}$Ni., Comment: Accepted for publication in the Astrophysical Journal (27 pages, 14 figures, 7 tables)

Estimating canopy height and its changes at meter resolution from satellite imagery is a significant challenge in computer vision with critical environmental applications. However, the lack of open-access datasets at this resolution hinders the reproducibility and evaluation of models. We introduce Open-Canopy, the first open-access, country-scale benchmark for very high-resolution (1.5 m) canopy height estimation, covering over 87,000 km$^2$ across France with 1.5 m resolution satellite imagery and aerial LiDAR data. Additionally, we present Open-Canopy-$\Delta$, a benchmark for canopy height change detection between images from different years at tree level-a challenging task for current computer vision models. We evaluate state-of-the-art architectures on these benchmarks, highlighting significant challenges and opportunities for improvement. Our datasets and code are publicly available at https://github.com/fajwel/Open-Canopy., Comment: 25 pages, 6+6 figures, Submitted to CVPR25

In this paper, performance of an Integrated Sensing and Communication (ISAC) Vehicle-to-Everything (V2X) scenario is evaluated, in which a vehicle simultaneously detects the next vehicle ahead while receiving a communication signal from a RoadSide Unit (RSU) of the infrastructure. Univariate and joint radar and communication performance metrics are evaluated within three different frameworks, namely the Stochastic Geometry (SG), Monte-Carlo (MC), and Ray-Tracing (RT) frameworks. The parameters of the system model are extracted from the RT simulations, and the metrics are compared to assess the accuracy of the SG framework. It is shown that the SG and MC system models are relevant w.r.t. RT simulations for the evaluation of univariate communication and sensing metrics, but larger discrepancies are observed for the joint metrics.

Reasoning about unpredicted change consists in explaining observations by events; we propose here an approach for explaining time-stamped observations by surprises, which are simple events consisting in the change of the truth value of a fluent. A framework for dealing with surprises is defined. Minimal sets of surprises are provided together with time intervals where each surprise has occurred, and they are characterized from a model-based diagnosis point of view. Then, a probabilistic approach of surprise minimisation is proposed.

The purpose of this study was to evaluate implementation of a whole school restorative practices model in Saint Paul Public Schools, Minnesota. Their model aims to embed practices to develop equitable relationships, engage students in learning, and respond relationally to behavioral concerns over a three-year period. The evaluation followed three cohorts of students in seven schools (4 elementary, 2 middle, and 1 high school) and included two quasi-experimental designed (QED) impact analyses and an analysis of implementation fidelity. The first QED was a school-level analysis focused on disproportionality in discipline outcomes, using a comparative short interrupted time series design with six treatment schools and a matched comparison group of 12 schools (N = 9,629 students). Outcome measures were school-level differences in dismissal and office discipline referral rates between White students and specific student groups of color (i.e., Black and Multiracial Black, Latinx, and American Indian/Alaskan Native students). The second QED was a longitudinal comparison of student-level attendance and discipline outcomes pre and post-implementation, with N= 5,043 students clustered within seven treatment schools and 14 matched comparison schools. Evaluation of fidelity was conducted on the seven schools that implemented restorative practices. Findings from the first QED showed that implementing restorative practices resulted in reduced disproportionality in differences in school-level dismissal rates between Black/Multiracial Black students and White students. The gap between Black and White student dismissals in treatment schools was reduced by an average of 0.021 dismissals per student more than the gap reduction observed in comparison schools (p = 0.02). No statistical evidence was found for differences between treatment and comparison schools for other analyses. For the second QED, no statistical support for improvements in student attendance and discipline was found, although the difference in dismissal rates between students attending restorative practice schools and comparison schools (1.6% vs. 1.0%) may have practical significance and is congruent with results from the first impact study. Fidelity results showed that treatment schools collectively met study thresholds for two of two core training components in year 1, one of four training and implementation components in year 2, and three of four training and implementation components in year 3. Higher levels of fidelity in year 3 indicate preliminary evidence of the minimum levels of program adherence needed to begin to impact school-level outcomes, such as reduced disproportionality in dismissal rates found in the first impact study. This study contributes promising evidence of whole school restorative practices' potential to reduce disparities in exclusionary discipline. Further research is needed to determine whether higher fidelity thresholds or longer periods of sustained implementation could have greater impacts on school-level and student outcomes. [This report was produced by the University of Minnesota's Healthy Youth Development -- Prevention Research Center.]

The goal of precision medicine is to provide individualized treatment at each stage of chronic diseases, a concept formalized by Dynamic Treatment Regimes (DTR). These regimes adapt treatment strategies based on decision rules learned from clinical data to enhance therapeutic effectiveness. Reinforcement Learning (RL) algorithms allow to determine these decision rules conditioned by individual patient data and their medical history. The integration of medical expertise into these models makes possible to increase confidence in treatment recommendations and facilitate the adoption of this approach by healthcare professionals and patients. In this work, we examine the mathematical foundations of RL, contextualize its application in the field of DTR, and present an overview of methods to improve its effectiveness by integrating medical expertise.

The purpose of this book is to provide an overview of AI research, ranging from basic work to interfaces and applications, with as much emphasis on results as on current issues. It is aimed at an audience of master students and Ph.D. students, and can be of interest as well for researchers and engineers who want to know more about AI. The book is split into three volumes.

The Near-InfraRed Planet Searcher or NIRPS is a precision radial velocity spectrograph developed through collaborative efforts among laboratories in Switzerland, Canada, Brazil, France, Portugal and Spain. NIRPS extends to the 0.98-1.8 $\mu$m domain of the pioneering HARPS instrument at the La Silla 3.6-m telescope in Chile and it has achieved unparalleled precision, measuring stellar radial velocities in the infrared with accuracy better than 1 m/s. NIRPS can be used either stand-alone or simultaneously with HARPS. Commissioned in late 2022 and early 2023, NIRPS embarked on a 5-year Guaranteed Time Observation (GTO) program in April 2023, spanning 720 observing nights. This program focuses on planetary systems around M dwarfs, encompassing both the immediate solar vicinity and transit follow-ups, alongside transit and emission spectroscopy observations. We highlight NIRPS's current performances and the insights gained during its deployment at the telescope. The lessons learned and successes achieved contribute to the ongoing advancement of precision radial velocity measurements and high spectral fidelity, further solidifying NIRPS' role in the forefront of the field of exoplanets., Comment: Proceeding at the SPIE Astronomical Telescopes + Instrumentation conference [Yokohama,Japan; June 2024]

The strong variations of energetic electron fluxes in the Earth's inner magnetosphere are notoriously hard to forecast. Developing accurate empirical models of electron fluxes from low to high altitudes at all latitudes is therefore useful to improve our understanding of flux variations and to assess radiation hazards for spacecraft systems. In the present work, energy- and pitch-angle-resolved precipitating, trapped, and backscattered electron fluxes measured at low altitude by Electron Loss and Fields Investigation (ELFIN) CubeSats are used to infer omnidirectional fluxes at altitudes below and above the spacecraft, from 150 km to 20,000 km, making use of adiabatic transport theory and quasi-linear diffusion theory. The inferred fluxes are fitted as a function of selected parameters using a stepwise multivariate optimization procedure, providing an analytical model of omnidirectional electron flux along each geomagnetic field line, based on measurements from only one spacecraft in low Earth orbit. The modeled electron fluxes are provided as a function of $L$-shell, altitude, energy, and two different indices of past substorm activity, computed over the preceding 4 hours or 3 days, potentially allowing to disentangle impulsive processes (such as rapid injections) from cumulative processes (such as inward radial diffusion and wave-driven energization). The model is validated through comparisons with equatorial measurements from the Van Allen Probes, demonstrating the broad applicability of the present method. The model indicates that both impulsive and time-integrated substorm activity partly control electron fluxes in the outer radiation belt and in the plasma sheet.

Transformers have revolutionized machine learning across diverse domains, yet understanding their behavior remains crucial, particularly in high-stakes applications. This paper introduces the contextual counting task, a novel toy problem aimed at enhancing our understanding of Transformers in quantitative and scientific contexts. This task requires precise localization and computation within datasets, akin to object detection or region-based scientific analysis. We present theoretical and empirical analysis using both causal and non-causal Transformer architectures, investigating the influence of various positional encodings on performance and interpretability. In particular, we find that causal attention is much better suited for the task, and that no positional embeddings lead to the best accuracy, though rotary embeddings are competitive and easier to train. We also show that out of distribution performance is tightly linked to which tokens it uses as a bias term.

Silicon Photomultipliers (SiPMs) are single photon detectors that gained increasing interest in many applications as an alternative to photomultiplier tubes. In particular in the field of space experiments, where volume, weight and power consumption are a major constraint, their advantages like compactness, ruggedness, and their potential to achieve high quantum efficiency from UV to NIR, makes them ideal candidates for spaceborne, low photon flux detectors. During space missions however, SiPMs are usually exposed to high levels of radiation, both ionizing and non-ionizing, which can deteriorate the performance of these detectors over time. The goal of this work is to compare process and layout variation of SiPMs in terms of their radiation damage effects to identify the features that helps reducing the deterioration of the performance and develop the next generation of more radiation tolerant detectors. To do this we irradiated with protons and X-rays several NUV-HD SiPMs with small area (single microcell, 0.2x0.2 mm2 and 1x1 mm2) produced in FBK. We performed online current-voltage measurements right after each irradiation step and a complete functional characterization before and after irradiation. We compare the results and show the most promising variations for future production of SiPMs for space applications.