As heterogeneous supercomputing architectures leveraging GPUs become increasingly central to high-performance computing (HPC), it is crucial for computational fluid dynamics (CFD) simulations, a de-facto HPC workload, to efficiently utilize such hardware. One of the key challenges of HPC codes is performance portability, i.e. the ability to maintain near-optimal performance across different accelerators. In the context of the REFMAP project, which targets scalable, GPU-enabled multi-fidelity CFD for urban airflow prediction, this paper analyzes the performance portability of SOD2D, a state-of-the-art Spectral Elements simulation framework across AMD and NVIDIA GPU architectures. We first discuss the physical and numerical models underlying SOD2D, highlighting its computational hotspots. Then, we examine its performance and scalability in a multi-level manner, i.e. defining and characterizing an extensive full-stack design space spanning across application, software and hardware infrastructure related parameters. Single-GPU performance characterization across server-grade NVIDIA and AMD GPU architectures and vendor-specific compiler stacks, show the potential as well as the diverse effect of memory access optimizations, i.e. 0.69× - 3.91× deviations in acceleration speedup. Performance variability of SOD2D at scale is further examined on the LUMI multi-GPU cluster, where profiling reveals similar throughput variations, highlighting the limits of performance projections and the need for multi-level, informed tuning.

High-dimensional unlabelled datasets present significant challenges for efficient analysis, storage and interpretation.Unsupervised feature selection offers a way to retain the most informative variables while discarding redundant or uninformative ones, enabling more scalable processing.We introduce a spatially aware, unsupervised method that uses information theoretic criteria to identify informative variables while limiting redundancy, producing compact and spatially dispersed subsets of features.Our approach avoids dependence on labelled data or modelspecific wrappers, making it suitable for large unstructured datasets.Experiments on MNIST and EMNIST datasets, including high-resolution upscaled versions, show that the selected features preserve both discriminative structure and reconstruction quality better than chosen supervised and unsupervised baselines, demonstrating the effectiveness of entropy and mutual information coupling in unlabelled high-dimensional settings.

This study presents a transformer-based approach for fault-tolerant control in fixed-wing Unmanned Aerial Vehicles (UAVs), designed to adapt in real time to dynamic changes caused by structural damage or actuator failures. Unlike traditional Flight Control Systems (FCSs) that rely on classical control theory and struggle under severe alterations in dynamics, our method directly maps outer-loop reference values (altitude, heading, and airspeed) into control commands using the in-context learning and attention mechanisms of transformers, thus bypassing inner-loop controllers and fault-detection layers. Employing a teacher-student knowledge distillation framework, the proposed approach trains a student agent with partial observations by transferring knowledge from a privileged expert agent with full observability, enabling robust performance across diverse failure scenarios. Experimental results demonstrate that our transformer-based controller outperforms industry-standard FCS and state-of-the-art reinforcement learning (RL) methods, maintaining high tracking accuracy and reliable flight behavior in nominal conditions and extreme failure cases, highlighting its potential for enhancing UAV operational safety and reliability. youtu.be/ATW3LZFRqc0

This study presents a comprehensive investigation of U-Net-based graph neural networks (Graph U-Nets) for mesh-agnostic spatio-temporal forecasting of unsteady flow fields. We systematically adapt and enhance Graph U-Nets, originally developed for classification tasks, for high-dimensional regression problems in fluid dynamics through extensive architectural modifications and hyperparameter optimization. Key enhancements include the implementation of Gaussian mixture model convolutional operators, which provide superior flexibility in modeling node dynamics and reduce prediction error by 95% compared to conventional graph convolutional operators. Additionally, we introduce noise injection strategies that significantly improve long-term prediction robustness, achieving an 86% reduction in temporal prediction error. Through comprehensive ablation studies, we investigate the effects of pooling strategies, normalization techniques, and architectural choices on model performance. We demonstrate the framework’s effectiveness in both transductive learning settings—successfully predicting flow fields in unseen spatial regions—and inductive learning scenarios across diverse mesh configurations with varying vortex shedding dynamics. Notably, we discover that optimal inductive performance requires eliminating pooling operations and employing layer normalization, contrary to single-mesh scenarios. The enhanced Graph U-Net successfully generalizes to unseen mesh scenarios, achieving improved prediction accuracy for challenging slow-vortex-shedding cases when trained on diverse flow regimes. This work establishes Graph U-Nets as a viable and flexible alternative to convolutional neural networks for computational fluid dynamics applications, demonstrating their potential for real-time flow prediction in digital twin frameworks across diverse industrial scenarios.

Abstract Multimodal Large Language Models (MLLMs) promise advanced vision-language capabilities, yet their effectiveness in visually presented mathematics remains underexplored. This paper analyses the development and evaluation of MLLMs for mathematical problem-solving, focusing on diagrams, multilingual text, and symbolic notation. The computational demands of evaluating these large-scale models across multilingual datasets necessitate high-performance computing infrastructure, as systematic benchmarking of state-of-the-art MLLMs requires distributed processing of thousands of inference requests and parallel evaluation across multiple model architectures. We then assess several models-including GPT-4o, Pixtral, Qwen-VL, Llama 3.2 Vision variants, and Gemini 2.0 Flash-in a multilingual Kangaroo-style benchmark spanning English, French, Spanish, and Catalan. Our experiments reveal four key findings. First, overall accuracy remains moderate across geometry, visual algebra, logic, patterns, and combinatorics: no single model excels in every topic. Second, whilst most models see improved accuracy with questions that do not have images, the gain is often limited; performance for some remains nearly unchanged without visual input, indicating underutilisation of diagrammatic information. Third, substantial variation exists across languages and difficulty levels: models frequently handle easier items but struggle with advanced geometry and combinatorial reasoning. Notably, Gemini 2.0 Flash achieves the highest accuracy on image-based tasks, followed by Qwen-VL 2.5 72B and GPT-4o, though none approach human-level performance. Fourth, a complementary analysis aimed at distinguishing whether models reason or simply recite reveals that Gemini and GPT-4o stand out for their structured reasoning and consistent accuracy. In contrast, Pixtral and Llama exhibit less consistent reasoning, often defaulting to heuristics or randomness when unable to align their outputs with the given answer options. Furthermore, detailed error analysis identifies two primary failure modes: encoding-stage errors, where models misidentify visual elements such as colours or shapes, and visio-semantic processing errors, where models struggle with three-dimensional spatial reasoning and geometric relationships, revealing systematic limitations even in state-of-the-art architectures.

As heterogeneous supercomputing architectures leveraging GPUs become increasingly central to high-performance computing (HPC), it is crucial for computational fluid dynamics (CFD) simulations to maintain performance portability.In this paper, we examine the performance and scalability of CFD framework SOD2D in a crosslayer manner, i.e. across application, software and hardware infrastructure related parameters.Single-GPU performance characterization across server-grade NVIDIA and AMD GPU architectures and vendor-specific compiler stacks, show the potential as well as the diverse effect of memory access optimizations, i.e. 0.69× -3.96× deviations in acceleration speedup.Performance variability of SOD2D at scale is then further examined on the LUMI multi-GPU cluster, showcasing analogous diverse effects on throughput, demonstrating the ineffectiveness of adopting performance projections, thus underscoring the importance and necessity of cross-layer informed performance analysis and tuning for multi-GPU configurations.

Are score function estimators a viable approach to learning with $k$-subset sampling? Sampling $k$-subsets is a fundamental operation that is not amenable to differentiable parametrization, impeding gradient-based optimization. Previous work has favored approximate pathwise gradients or relaxed sampling, dismissing score function estimators because of their high variance. Inspired by the success of score function estimators in variational inference and reinforcement learning, we revisit them for $k$-subset sampling. We demonstrate how to efficiently compute the distribution's score function using a discrete Fourier transform and reduce the estimator's variance with control variates. The resulting estimator provides both $k$-hot samples and unbiased gradient estimates while being applicable to non-differentiable downstream models, unlike existing methods. We validate our approach experimentally and find that it produces results comparable to those of recent state-of-the-art pathwise gradient estimators across a range of tasks.

Inverse problems have many applications in science and engineering. In Computer vision, several image restoration tasks such as inpainting, deblurring, and super-resolution can be formally modeled as inverse problems. Recently, methods have been developed for solving inverse problems that only leverage a pre-trained unconditional diffusion model and do not require additional task-specific training. In such methods, however, the inherent intractability of determining the conditional score function during the reverse diffusion process poses a real challenge, leaving the methods to settle with an approximation instead, which affects their performance in practice. Here, we propose a MAP estimation framework to model the reverse conditional generation process of a continuous time diffusion model as an optimization process of the underlying MAP objective, whose gradient term is tractable. In theory, the proposed framework can be applied to solve general inverse problems using gradient-based optimization methods. However, given the highly non-convex nature of the loss objective, finding a perfect gradient-based optimization algorithm can be quite challenging, nevertheless, our framework offers several potential research directions. We use our proposed formulation to develop empirically effective algorithms for image restoration. We validate our proposed algorithms with extensive experiments over multiple datasets across several restoration tasks.

There is much debate on whether quantum computing on current NISQ devices, consisting of noisy hundred qubits and requiring a non-negligible usage of classical computing as part of the algorithms, has utility and will ever offer advantages for scientific and industrial applications with respect to traditional computing. In this position paper, we argue that while real-world NISQ quantum applications have yet to surpass their classical counterparts, strategic approaches can be used to facilitate advancements in both industrial and scientific applications. We have identified three key strategies to guide NISQ computing towards practical and useful implementations. Firstly, prioritizing the identification of a "killer app" is a key point. An application demonstrating the distinctive capabilities of NISQ devices can catalyze broader development. We suggest focusing on applications that are inherently quantum, e.g., pointing towards quantum chemistry and material science as promising domains. These fields hold the potential to exhibit benefits, setting benchmarks for other applications to follow. Secondly, integrating AI and deep-learning methods into NISQ computing is a promising approach. Examples such as quantum Physics-Informed Neural Networks and Differentiable Quantum Circuits (DQC) demonstrate the synergy between quantum computing and AI. Lastly, recognizing the interdisciplinary nature of NISQ computing, we advocate for a co-design approach. Achieving synergy between classical and quantum computing necessitates an effort in co-designing quantum applications, algorithms, and programming environments, and the integration of HPC with quantum hardware. The interoperability of these components is crucial for enabling the full potential of NISQ computing. In conclusion, through the usage of these three approaches, we argue that NISQ computing can surpass current limitations and evolve into a valuable tool for scientific and industrial applications. This requires an approach that integrates domain-specific killer apps, harnesses the power of quantum-enhanced AI, and embraces a collaborative co-design methodology.

The aviation field is rapidly evolving towards an era where both typical aviation and Unmanned Aicraft Systems are essential and co-exist in the same airspace. This new territory raises important concerns regarding environmental impact, safety and societal acceptance. The RefMap European Project is an initiative that addresses these issues and aims at optimizing air traffic in terms of the environmental footprint in aviation and drone flights. One of RefMap's objectives is the development of powerful deep-learning models that predict urban flow based on extensive CFD simulations. The excessive time requirements of CFD simulations require the computational power of exascale heterogeneous supercomputer clusters. This work presents RefMap's strategy to mitigate simulation to GPU-enabled high-class solvers and further leverage sophisticated autotuning HPC techniques for creating portable high-performance simulations that can efficiently run on any GPU architecture and parallel system.

Artificial Intelligence (AI) has the potential to revolutionize space exploration by delegating spacecraft decisions to onboard AI. The onboard neural-network (NN) will have parameters that can be updated onboard by telecommands after training on ground. However, Satellite uplinks have limited bandwidth and transmissions can be costly. Furthermore, a suboptimal NN will miss valuable scientific data. Smaller networks can therefore decrease the uplink cost and increase the value of the downlinked data. In this work, we evaluate and discuss using reduced-precision and small NNs to reduce the upload time. As an example of a mission where AI could be used, we focus on NASA’s Magnetospheric MultiScale (MMS) mission. We showcase how an onboard AI can be used in the Earth’s magnetosphere to classify data for selective downlink or recognize a region of interest to trigger a burst-mode, collecting data at a high-rate. Using a simple algorithm, we show the detection of a region of interest in on a stream of classifications. To provide the classifications, we use a Convolutional Neural Network (CNN) trained to an accuracy >94%. We show how the NN can be reduced to a single linear layer without accuracy loss. Thereby, Reducing the upload time by up to 98.9%. Each network can be reduced further by using lower-precision formats, changing the accuracy by less than 0.6 percentage points.

Feature selection is a crucial task in settings where data is high-dimensional or acquiring the full set of features is costly. Recent developments in neural network-based embedded feature selection show promising results across a wide range of applications. Concrete Autoencoders (CAEs), considered state-of-the-art in embedded feature selection, may struggle to achieve stable joint optimization, hurting their training time and generalization. In this work, we identify that this instability is correlated with the CAE learning duplicate selections. To remedy this, we propose a simple and effective improvement: Indirectly Parameterized CAEs (IP-CAEs). IP-CAEs learn an embedding and a mapping from it to the Gumbel-Softmax distributions’ parameters. Despite being simple to implement, IP-CAE exhibits significant and consistent improvements over CAE in both generalization and training time across several datasets for reconstruction and classification. Unlike CAE, IP-CAE effectively leverages non-linear relationships and does not require retraining the jointly optimized decoder. Furthermore, our approach is, in principle, generalizable to Gumbel-Softmax distributions beyond feature selection.

Thunderstorms can be a large source of disruption for European air-traffic management causing a chaotic state of operation within the airspace system. In current practice, air-traffic managers are provided with imprecise forecasts which limit their ability to plan strategically. As a result, weather mitigation is performed using tactical measures with a time horizon of three hours. Increasing the lead time of thunderstorm predictions to the day before operations could help air-traffic managers plan around weather and improve the efficiency of air-traffic-management operations. Emerging techniques based on machine learning have provided promising results, partly attributed to reduced human bias and improved capacity in predicting thunderstorms purely from numerical weather prediction data. In this paper, we expand on our previous work on thunderstorm forecasting, by applying convolutional neural networks (CNNs) to exploit the spatial characteristics embedded in the weather data. The learning task of predicting convection is formulated as a binary-classification problem based on satellite data. The performance of multiple CNN-based architectures, including a fully-convolutional neural network (FCN), a CNN-based encoder–decoder, a U-Net, and a pyramid-scene parsing network (PSPNet) are compared against a multi-layer-perceptron (MLP) network. Our work indicates that CNN-based architectures improve the performance of point-prediction models, with a fully-convolutional neural-network architecture having the best performance. Results show that CNN-based architectures can be used to increase the prediction lead time of thunderstorms. Lastly, a case study illustrating the applications of convection-prediction models in an air-traffic-management setting is presented.

This study explores the topics and trends of teaching AI ethics in higher education, using Latent Dirichlet Allocation as the analysis tool. The analyses included 166 courses from 105 universities around the world. Building on the uncovered patterns, we distil a model of current pedagogical practice, the BAG model (Build, Assess, and Govern), that combines cognitive levels, course content, and disciplines. The study critically assesses the implications of this teaching paradigm and challenges practitioners to reflect on their practices and move beyond stereotypes and biases.

Modal-decomposition techniques are computational frameworks based on data aimed at identifying a low-dimensional space for capturing dominant flow features: the so-called modes. We propose a deep probabilistic-neural-network architecture for learning a minimal and near-orthogonal set of non-linear modes from high-fidelity turbulent-flow data useful for flow analysis, reduced-order modeling and flow control. Our approach is based on β-variational autoencoders (β-VAEs) and convolutional neural networks (CNNs), which enable extracting non-linear modes from multi-scale turbulent flows while encouraging the learning of independent latent variables and penalizing the size of the latent vector. Moreover, we introduce an algorithm for ordering VAE-based modes with respect to their contribution to the reconstruction. We apply this method for non-linear mode decomposition of the turbulent flow through a simplified urban environment, where the flow-field data is obtained based on well-resolved large-eddy simulations (LESs). We demonstrate that by constraining the shape of the latent space, it is possible to motivate the orthogonality and extract a set of parsimonious modes sufficient for high-quality reconstruction. Our results show the excellent performance of the method in the reconstruction against linear-theory-based decompositions, where the energy percentage captured by the proposed method from five modes is equal to 87.36% against 32.41% of the POD. Moreover, we compare our method with available AE-based models. We show the ability of our approach in the extraction of near-orthogonal modes with the determinant of the correlation matrix equal to 0.99, which may lead to interpretability.

The domain of Artificial Intelligence (AI) ethics is not new, with discussions going back at least 40 years. Teaching the principles and requirements of ethical AI to students is considered an essential part of this domain, with an increasing number of technical AI courses taught at several higher-education institutions around the globe including content related to ethics. By using Latent Dirichlet Allocation (LDA), a generative probabilistic topic model, this study uncovers topics in teaching ethics in AI courses and their trends related to where the courses are taught, by whom, and at what level of cognitive complexity and specificity according to Bloom’s taxonomy. In this exploratory study based on unsupervised machine learning, we analyzed a total of 166 courses: 116 from North American universities, 11 from Asia, 36 from Europe, and 10 from other regions. Based on this analysis, we were able to synthesize a model of teaching approaches, which we call BAG (Build, Assess, and Govern), that combines specific cognitive levels, course content topics, and disciplines affiliated with the department(s) in charge of the course. We critically assess the implications of this teaching paradigm and provide suggestions about how to move away from these practices. We challenge teaching practitioners and program coordinators to reflect on their usual procedures so that they may expand their methodology beyond the confines of stereotypical thought and traditional biases regarding what disciplines should teach and how. This article appears in the AI & Society track.

We are pleased to present a special issue that focuses on the software systems for managing the impact of the coronavirus disease 19 (COVID-19) pandemic. The COVID-19 pandemic has affected around 192 million people worldwide and has led to ˜4.13 million deaths as of July 22, 2021. Globally, most of the countries have implemented lockdowns to protect their citizens. However, lockdown over an extended period is unsustainable. Hence, it is widely believed that virus testing and tracking is the best approach to ease lockdown measures. There is a need for innovative software systems to manage the impact of the COVID-19 pandemic effectively in many areas such as healthcare system, transport systems, supply-chain system, educational system, government-service delivery, pharmaceutical companies, manufacturing, software industries, and multinational companies. For example, in healthcare, smart-software systems would be able to remotely measure a person's body temperature, heart and respiratory rates, identifying their movements (including sneezing, coughing, shivering, etc.) to identify whether a person is displaying symptoms of COVID-19 or not. An essential aspect associated with these technologies is data privacy, scalability, and quality of service (QoS) in terms of reliability, availability, security, latency, and energy which need to be considered throughout the development of the software systems. In countries like India, UK, Russia, Brazil, and USA, the system would also help to ensure that isolated communities have access to testing, delivered in a fast, accurate, and efficient manner. These software systems would help and support the assessment of public-health strategies and policies such as social distancing and assess further interventions to control the spread of the virus. Innovative software systems can increase stakeholder participation, as cost-effective assistance in the COVID-19 pandemic monitoring is of great interest to many countries. To manage the impact of this pandemic, there is a need to design and develop scalable, reliable, and energy-efficient sustainable software solutions for different COVID-19 scenarios. In consideration of the existing systems and their features, an Internet of Things (IoT)-based system suitable for COVID-19 or pandemic situations associated with other influenza viruses can be developed. Furthermore, these systems can be integrated with artificial-intelligence (AI) processes for effective data-collection, analysis, statistical visualization, sharing, and decision making. Moreover, these systems can be implemented using both simulations and real-time testbeds for COVID-19 operations (sanitization, medication, monitoring, thermal imaging, etc.) to test their performance in terms of scalability, reliability, availability, and energy efficiency. There is a need to use AI methods, such as reinforcement learning, deep learning, and genetic algorithms while developing IoT-based software systems to achieve self-learning, self-adaptation, and autonomous decision-making capabilities in order to improve efficiency of the systems. Meanwhile, a huge voluminous amount of complex data is generated from various sources including World Health Organization (WHO), social networking, edge devices, private and public hospitals, patients and academic institutes, which needs an effective big data analytics mechanism to manage this data proficiently. Furthermore, there is a need to study the impact of system configuration on workload processing at different cloud nodes while maintaining the QoS dynamically. The data are collected in databases, it is subsequently examined and monitored, and it is important to manage data consistency and integrity. In this context, we argue that it is essential to employ decentralized data-gathering approaches, maintaining the privacy of the population as a high priority. This special issue has received articles by researchers and practitioners from both academia and industry to develop innovative software systems for managing the impact of the COVID-19 pandemic. This special issue, therefore, aims to focus the attention of its readers to four research articles carefully selected after multiple rounds of peer-review. The brief contributions of these papers are discussed in the following section: The first paper entitled "An approach to forecast impact of COVID-19 using supervised machine learning model" by Mohan et al.1 proposes a hybrid model to predict the effect of COVID-19 using moving regressive, autoregressive, and ensemble learning model. This work uses two datasets from Worldometer and Ministry of Health & Family Welfare of India to conduct the countrywise predictions across the world and statewise predictions of India, respectively. The second paper entitled "NovidChain: Blockchain-based privacy-preserving platform for COVID-19 test/vaccine certificates" by Abid et al.2 includes various promising ideas such as maintains the immutability and data integrity using Blockchain technology, enhances the privacy by incorporating encryption for personal information and verifies the COVID-19 proof using W3C verifiable credentials standard immediately. The third paper entitled "Software System to Predict the Infection in COVID-19 Patients using Deep Learning and Web of Things" by Singh et al.3 generates synthetic data using various data augmentation techniques. Proposed system uses U-Net and WoT to segment the COVID Medseg and Radiopedia datasets in an autonomic manner. Experimental results show that the system gives better performance in terms of network latency, response time, and server latency. The fourth paper entitled "Advanced Data Integration in Banking, Financial, and Insurance Software in the Age of COVID-19" by Maiti et al.4 contributes to recognize the effect of the COVID-19 pandemic on the global Banking Financial Services and Insurance landscape. Further, a hype cycle has been developed to find out the important software technologies to handle real-world challenges related to corporate. We believe the work that has been approved in this special issue will assist readers of the journal and a broader research community to learn about the topics of software systems and impacts of COVID-19 pandemic, and inspire them to study more in this area. We would like to express our gratitude to the Editor-in-Chief (Prof. Rajkumar Buyya) and editorial board members for allowing us to bring out this special issue and guiding us throughout the process. We also want to express our gratitude to and further acknowledge the administrative staff, reviewers, and especially the authors for their contributions to the success of this issue.

Abstract In situ visualization on high-performance computing systems allows us to analyze simulation results that would otherwise be impossible, given the size of the simulation data sets and offline post-processing execution time. We develop an in situ adaptor for Paraview Catalyst and Nek5000, a massively parallel Fortran and C code for computational fluid dynamics. We perform a strong scalability test up to 2048 cores on KTH’s Beskow Cray XC40 supercomputer and assess in situ visualization’s impact on the Nek5000 performance. In our study case, a high-fidelity simulation of turbulent flow, we observe that in situ operations significantly limit the strong scalability of the code, reducing the relative parallel efficiency to only $$\approx 21\%$$ ≈ 21 % on 2048 cores (the relative efficiency of Nek5000 without in situ operations is $$\approx 99\%$$ ≈ 99 % ). Through profiling with Arm MAP, we identified a bottleneck in the image composition step (that uses the Radix-kr algorithm) where a majority of the time is spent on MPI communication. We also identified an imbalance of in situ processing time between rank 0 and all other ranks. In our case, better scaling and load-balancing in the parallel image composition would considerably improve the performance of Nek5000 with in situ capabilities. In general, the result of this study highlights the technical challenges posed by the integration of high-performance simulation codes and data-analysis libraries and their practical use in complex cases, even when efficient algorithms already exist for a certain application scenario.