Search

We present learning-based implicit shape representations designed for real-time avatar collision queries arising in the simulation of clothing. Signed distance functions (SDFs) have been used for such queries for many years due to their computational efficiency. Recently deep neural networks have been used for implicit shape representations (DeepSDFs) due to their ability to represent multiple shapes with modest memory requirements compared to traditional representations over dense grids. However, the computational expense of DeepSDFs prevents their use in real-time clothing simulation applications. We design a learning-based representation of SDFs for human avatars whoes bodies change shape kinematically due to joint-based skinning. Rather than using a single DeepSDF for the entire avatar, we use a collection of extremely computationally efficient (shallow) neural networks that represent localized deformations arising from changes in body shape induced by the variation of a single joint. This requires a stitching process to combine each shallow SDF in the collection together into one SDF representing the signed closest distance to the boundary of the entire body. To achieve this we augment each shallow SDF with an additional output that resolves whether or not the individual shallow SDF value is referring to a closest point on the boundary of the body, or to a point on the interior of the body (but on the boundary of the individual shallow SDF). Our model is extremely fast and accurate and we demonstrate its applicability with real-time simulation of garments driven by animated characters., Comment: Preprint

Federated learning has recently gained popularity as a framework for distributed clients to collaboratively train a machine learning model using local data. While traditional federated learning relies on a central server for model aggregation, recent advancements adopt a decentralized framework, enabling direct model exchange between clients and eliminating the single point of failure. However, existing decentralized frameworks often assume all clients train a shared model. Personalizing each client's model can enhance performance, especially with heterogeneous client data distributions. We propose FedSPD, an efficient personalized federated learning algorithm for the decentralized setting, and show that it learns accurate models even in low-connectivity networks. To provide theoretical guarantees on convergence, we introduce a clustering-based framework that enables consensus on models for distinct data clusters while personalizing to unique mixtures of these clusters at different clients. This flexibility, allowing selective model updates based on data distribution, substantially reduces communication costs compared to prior work on personalized federated learning in decentralized settings. Experimental results on real-world datasets show that FedSPD outperforms multiple decentralized variants of personalized federated learning algorithms, especially in scenarios with low-connectivity networks.

The High Temperature Reactor Code Package provides sophisticated modeling and simulation capabilities for high temperature gas cooled reactors like the HTR-200 Modul. However, HCP currently lacks integrated methods for uncertainty quantification and sensitivity analysis. This work aims to couple HCP with the DAKOTA toolkit to enable UQ workflows for quantifying how different uncertainties impact HTGR system performance. DAKOTA offers state of the art sampling and analysis methods that will be linked to the HCP time-dependent multiphysics environment. Key input parameters related to manufacturing variability, boundary and initial conditions, and material properties will be defined as uncertain in this study. Both steady state and time-dependent multiphysics simulations will be analyzed to understand the relative importance of uncertainties across different physics phenomena. Output metrics of interest can include anything measured by the HCP code. Results show the HTR-200 Modul design's robustness to input uncertainties related to inlet gas temperature, U-235 enrichment, graphite density, inlet mass flow rate, and reactor power. A pressurized loss of forced cooling transient was simulated against uncertainty in the inlet gas temperature. Results show the maximum fuel temperature is within design limits with a relatively big safety margin even when considering uncertainty in the boundary conditions of the reactor. Future work will expand the analysis to more HCP physics modules and nuclear data uncertainties. By leveraging UQ and sensitivity analysis with the linked HCP/DAKOTA environment, this work will provide valuable new insights into the expected performance variability of HTR systems during normal and off-normal conditions. The framework developed here will aid uncertainty management activities by gas-cooled reactor developers, researchers, and regulators.

We introduce a dynamic spatiotemporal volatility model that extends traditional approaches by incorporating spatial, temporal, and spatiotemporal spillover effects, along with volatility-specific observed and latent factors. The model offers a more general network interpretation, making it applicable for studying various types of network spillovers. The primary innovation lies in incorporating volatility-specific latent factors into the dynamic spatiotemporal volatility model. Using Bayesian estimation via the Markov Chain Monte Carlo (MCMC) method, the model offers a robust framework for analyzing the spatial, temporal, and spatiotemporal effects of a log-squared outcome variable on its volatility. We recommend using the deviance information criterion (DIC) and a regularized Bayesian MCMC method to select the number of relevant factors in the model. The model's flexibility is demonstrated through two applications: a spatiotemporal model applied to the U.S. housing market and another applied to financial stock market networks, both highlighting the model's ability to capture varying degrees of interconnectedness. In both applications, we find strong spatial/network interactions with relatively stronger spillover effects in the stock market.

We present a neural network-enhanced column generation (CG) approach for a parallel machine scheduling problem. The proposed approach utilizes an encoder-decoder attention model, namely the transformer and pointer architectures, to develop job sequences with negative reduced cost and thus generate columns to add to the master problem. By training the neural network offline and using it in inference mode to predict negative reduced costs columns, we achieve significant computational time savings compared to dynamic programming (DP). Since the exact DP procedure is used to verify that no further columns with negative reduced cost can be identified at termination, the optimality guarantee of the original CG procedure is preserved. For small to medium-sized instances, our approach achieves an average 45% reduction in computation time compared to solving the subproblems with DP. Furthermore, the model generalizes not only to unseen, larger problem instances from the same probability distribution but also to instances from different probability distributions than those presented at training time. For large-sized instances, the proposed approach achieves an 80% improvement in the objective value in under 500 seconds, demonstrating both its scalability and efficiency.

High-quality spatiotemporal traffic data is crucial for intelligent transportation systems (ITS) and their data-driven applications. Inevitably, the issue of missing data caused by various disturbances threatens the reliability of data acquisition. Recent studies of diffusion probability models have demonstrated the superiority of deep generative models in imputation tasks by precisely capturing the spatio-temporal correlation of traffic data. One drawback of diffusion models is their slow sampling/denoising process. In this work, we aim to accelerate the imputation process while retaining the performance. We propose a fast conditional diffusion model for spatiotemporal traffic data imputation (FastSTI). To speed up the process yet, obtain better performance, we propose the application of a high-order pseudo-numerical solver. Our method further revs the imputation by introducing a predefined alignment strategy of variance schedule during the sampling process. Evaluating FastSTI on two types of real-world traffic datasets (traffic speed and flow) with different missing data scenarios proves its ability to impute higher-quality samples in only six sampling steps, especially under high missing rates (60\% $\sim$ 90\%). The experimental results illustrate a speed-up of $\textbf{8.3} \times$ faster than the current state-of-the-art model while achieving better performance., Comment: This paper has been accepted by IEEE Transactions on Intelligent Transportation Systems for publication. Permission from IEEE must be obtained for all other uses, in any current or future media

To adapt to real-world dynamics, intelligent systems need to assimilate new knowledge without catastrophic forgetting, where learning new tasks leads to a degradation in performance on old tasks. To address this, continual learning concept is proposed for enabling autonomous systems to acquire new knowledge and dynamically adapt to changing environments. Specifically, energy-efficient continual learning is needed to ensure the functionality of autonomous systems under tight compute and memory resource budgets (i.e., so-called autonomous embedded systems). Neuromorphic computing, with brain-inspired Spiking Neural Networks (SNNs), offers inherent advantages for enabling low-power/energy continual learning in autonomous embedded systems. In this paper, we comprehensively discuss the foundations and methods for enabling continual learning in neural networks, then analyze the state-of-the-art works considering SNNs. Afterward, comparative analyses of existing methods are conducted while considering crucial design factors, such as network complexity, memory, latency, and power/energy efficiency. We also explore the practical applications that can benefit from SNN-based continual learning and open challenges in real-world scenarios. In this manner, our survey provides valuable insights into the recent advancements of SNN-based continual learning for real-world application use-cases., Comment: This work has been submitted to the IEEE Access for possible publication

The Greenland Ice Sheet (GrIS) has emerged as a significant contributor to global sea level rise, primarily due to increased meltwater runoff. Supraglacial lakes, which form on the ice sheet surface during the summer months, can impact ice sheet dynamics and mass loss; thus, better understanding these lakes' seasonal evolution and dynamics is an important task. This study presents a computationally efficient time series classification approach that uses Gaussian Mixture Models (GMMs) of the Reconstructed Phase Spaces (RPSs) to identify supraglacial lakes based on their seasonal evolution: 1) those that refreeze at the end of the melt season, 2) those that drain during the melt season, and 3) those that become buried, remaining liquid insulated a few meters beneath the surface. Our approach uses time series data from the Sentinel-1 and Sentinel-2 satellites, which utilize microwave and visible radiation, respectively. Evaluated on a GrIS-wide dataset, the RPS-GMM model, trained on a single representative sample per class, achieves 85.46% accuracy with Sentinel-1 data alone and 89.70% with combined Sentinel-1 and Sentinel-2 data. This performance significantly surpasses existing machine learning and deep learning models which require a large training data. The results demonstrate the robustness of the RPS-GMM model in capturing the complex temporal dynamics of supraglacial lakes with minimal training data., Comment: Accepted for publication in 23rd International Conference on Machine Learning and Applications (ICMLA 2024)

In an increasingly interconnected world, a key scientific challenge is to examine mechanisms that lead to the widespread propagation of contagions, such as misinformation and pathogens, and identify risk factors that can trigger large-scale outbreaks. Underlying both the spread of disease and misinformation epidemics is the evolution of the contagion as it propagates, leading to the emergence of different strains, e.g., through genetic mutations in pathogens and alterations in the information content. Recent studies have revealed that models that do not account for heterogeneity in transmission risks associated with different strains of the circulating contagion can lead to inaccurate predictions. However, existing results on multi-strain spreading assume that the network has a vanishingly small clustering coefficient, whereas clustering is widely known to be a fundamental property of real-world social networks. In this work, we investigate spreading processes that entail evolutionary adaptations on random graphs with tunable clustering and arbitrary degree distributions. We derive a mathematical framework to quantify the epidemic characteristics of a contagion that evolves as it spreads, with the structure of the underlying network as given via arbitrary {\em joint} degree distributions of single-edges and triangles. To the best of our knowledge, our work is the first to jointly analyze the impact of clustering and evolution on the emergence of epidemic outbreaks. We supplement our theoretical finding with numerical simulations and case studies, shedding light on the impact of clustering on contagion spread.

The realization of fault-tolerant quantum computing requires the execution of quantum error-correction (QEC) schemes, to mitigate the fragile nature of qubits. In this context, to ensure the success of QEC, a protocol capable of implementing both qubit reset and leakage reduction is highly desirable. We demonstrate such a protocol in an architecture consisting of fixed-frequency transmon qubits pair-wise coupled via tunable couplers -- an architecture that is compatible with the surface code. We use tunable couplers to transfer any undesired qubit excitation to the readout resonator of the qubit, from which this excitation decays into the feedline. In total, the combination of qubit reset, leakage reduction, and coupler reset takes only 83ns to complete. Our reset scheme is fast, unconditional, and achieves fidelities well above 99%, thus enabling fixed-frequency qubit architectures as future implementations of fault-tolerant quantum computers. Our protocol also provides a means to both reduce QEC cycle runtime and improve algorithmic fidelity on quantum computers.

We prove the universality of the joint distribution of an eigenvalue and the corresponding diagonal eigenvector overlap, in the bulk and at the edge, for eigenvalues of complex matrices and real eigenvalues of real matrices. As part of the proof we obtain a bound for the least non-zero singular value of $X-z$ when $z$ is an edge eigenvalue and a bound for the inner product between left and right singular vectors of $X-z$ when $|z|=1+O(N^{-1/2})$.

In smart retail applications, the large number of products and their frequent turnover necessitate reliable zero-shot object classification methods. The zero-shot assumption is essential to avoid the need for re-training the classifier every time a new product is introduced into stock or an existing product undergoes rebranding. In this paper, we make three key contributions. Firstly, we introduce the MIMEX dataset, comprising 28 distinct product categories. Unlike existing datasets in the literature, MIMEX focuses on fine-grained product classification and includes a diverse range of retail products. Secondly, we benchmark the zero-shot object classification performance of state-of-the-art vision-language models (VLMs) on the proposed MIMEX dataset. Our experiments reveal that these models achieve unsatisfactory fine-grained classification performance, highlighting the need for specialized approaches. Lastly, we propose a novel ensemble approach that integrates embeddings from CLIP and DINOv2 with dimensionality reduction techniques to enhance classification performance. By combining these components, our ensemble approach outperforms VLMs, effectively capturing visual cues crucial for fine-grained product discrimination. Additionally, we introduce a class adaptation method that utilizes visual prototyping with limited samples in scenarios with scarce labeled data, addressing a critical need in retail environments where product variety frequently changes. To encourage further research into zero-shot object classification for smart retail applications, we will release both the MIMEX dataset and benchmark to the research community. Interested researchers can contact the authors for details on the terms and conditions of use. The code is available: https://github.com/AnilOsmanTur/Zero-shot-Retail-Product-Classification., Comment: Accepted at 2024 IEEE 8th Forum on Research and Technologies for Society and Industry Innovation (RTSI) conference

We prove an effective slope gap distribution result first for the square torus and then for general lattice translation surfaces. As a corollary, we obtain a dynamical proof for an effective gap distribution result for the Farey fractions. As an intermediate step, we prove an effective equidistribution result for the intersection points of long horocycles with a particular transversal of the horocycle flow in $\mathrm{SL}_2 (\mathbb R)/\Gamma$ where $\Gamma$ is a lattice., Comment: 37 pages, 3 figures

Reproducibility is essential for scientific research. However, in computer vision, achieving consistent results is challenging due to various factors. One influential, yet often unrecognized, factor is CUDA-induced randomness. Despite CUDA's advantages for accelerating algorithm execution on GPUs, if not controlled, its behavior across multiple executions remains non-deterministic. While reproducibility issues in ML being researched, the implications of CUDA-induced randomness in application are yet to be understood. Our investigation focuses on this randomness across one standard benchmark dataset and two real-world datasets in an isolated environment. Our results show that CUDA-induced randomness can account for differences up to 4.77% in performance scores. We find that managing this variability for reproducibility may entail increased runtime or reduce performance, but that disadvantages are not as significant as reported in previous studies.

This study introduces a novel approach for crystal structure analysis, utilizing Inverse EXAFS Analysis (IEA). To assess the reliability of IEA, we applied it to various experimentally studied materials, including LiCrO2 and CuFeO2. Our findings demonstrate that IEA offers a promising alternative to traditional techniques like XRD, particularly in cases where instrumentation or crystal structure defects pose challenges. IEA effectively revealed the crystal structures of both LiCrO2 and CuFeO2, demonstrating its ability to accurately characterize complex materials. The technique's potential to enhance XAFS data analysis is significant, providing researchers with a valuable tool for crystal structure determination. Future developments in IEA could further expand its capabilities and make it a more accessible and efficient method for materials scientists.

In this paper, we jointly combine image classification and image denoising, aiming to enhance human perception of noisy images captured by edge devices, like low-light security cameras. In such settings, it is important to retain the ability of humans to verify the automatic classification decision and thus jointly denoise the image to enhance human perception. Since edge devices have little computational power, we explicitly optimize for efficiency by proposing a novel architecture that integrates the two tasks. Additionally, we alter a Neural Architecture Search (NAS) method, which searches for classifiers to search for the integrated model while optimizing for a target latency, classification accuracy, and denoising performance. The NAS architectures outperform our manually designed alternatives in both denoising and classification, offering a significant improvement to human perception. Our approach empowers users to construct architectures tailored to domains like medical imaging, surveillance systems, and industrial inspections., Comment: Accepted paper at the ECCV 2024 workshop on Advances in Image Manipulation (AIM)

Spin-photon interfaces based on group-IV colour centres in diamond offer a promising platform for quantum networks. A key challenge in the field is realizing precise single-defect positioning and activation, which is crucial for scalable device fabrication. Here we address this problem by demonstrating a two-step fabrication method for tin vacancy (SnV-) centres that uses site-controlled ion implantation followed by local femtosecond laser annealing with in-situ spectral monitoring. The ion implantation is performed with sub-50 nm resolution and a dosage that is controlled from hundreds of ions down to single ions per site, limited by Poissonian statistics. Using this approach, we successfully demonstrate site-selective creation and modification of single SnV- centres. The technique opens a window onto materials tuning at the single defect level, and provides new insight into defect structures and dynamics during the annealing process. While demonstrated for SnV- centres, this versatile approach can be readily generalised to other implanted colour centres in diamond and wide-bandgap materials., Comment: 10 pages, 4 figures, methods, and supplementary information

Proton beam radiotherapy stands at the forefront of precision cancer treatment, leveraging the unique physical interactions of proton beams with human tissue to deliver minimal dose upon entry and deposit the therapeutic dose precisely at the so-called Bragg peak, with no residual dose beyond this point. The Bragg peak is the characteristic maximum that occurs when plotting the curve describing the rate of energy deposition along the length of the proton beam. Moreover, as a natural phenomenon, it is caused by an increase in the rate of nuclear interactions of protons as their energy decreases. From an analytical perspective, Bortfeld proposed a parametric family of curves that can be accurately calibrated to data replicating the Bragg peak in one dimension. We build, from first principles, the very first mathematical model describing the energy deposition of protons. Our approach uses stochastic differential equations and affords us the luxury of defining the natural analogue of the Bragg curve in two or three dimensions. This work is purely theoretical and provides a new mathematical framework which is capable of encompassing models built using Geant4 Monte Carlo, at one extreme, to pencil beam calculations with Bortfeld curves at the other.

We present the first homogeneous release of several thousand Type Ia supernovae (SNe Ia), all having spectroscopic classification, and spectroscopic redshifts for half the sample. This release, named the "DR2", contains 3628 nearby (z < 0.3) SNe Ia discovered, followed and classified by the Zwicky Transient Facility survey between March 2018 and December 2020. Of these, 3000 have good-to-excellent sampling and 2667 pass standard cosmology light-curve quality cuts. This release is thus the largest SN Ia release to date, increasing by an order of magnitude the number of well characterized low-redshift objects. With the "DR2", we also provide a volume-limited (z < 0.06) sample of nearly a thousand SNe Ia. With such a large, homogeneous and well controlled dataset, we are studying key current questions on SN cosmology, such as the linearity SNe Ia standardization, the SN and host dependencies, the diversity of the SN Ia population, and the accuracy of the current light-curve modeling. These, and more, are studied in detail in a series of articles associated with this release. Alongside the SN Ia parameters, we publish our force-photometry gri-band light curves, 5138 spectra, local and global host properties, observing logs, and a python tool to ease use and access of these data. The photometric accuracy of the "DR2" is not yet suited for cosmological parameter inference, which will follow as "DR2.5" release. We nonetheless demonstrate that the multi-thousand SN Ia Hubble Diagram has a typical 0.15 mag scatter., Comment: ZTF SN Ia DR2 release paper. Submitted to A&A (ZTF DR2 Special Issue). Already 1 response to referee

We characterise a small footprint travelling-wave parametric amplifier (TWPA). The TWPA is built with magnetically flux-tunable superconducting nonlinear asymmetric inductive elements (SNAILs) and parallel-plate capacitors. It implements three-wave mixing (3WM) with resonant phase matching (RPM), a small cutoff frequency for high gain per unitcell and impedance matching networks for large bandwidth impedance matching. The device has 200 unitcells and a physical footprint of only 1.1 mm^2, yet demonstrates an average parametric gain of 19 dB over a 3 GHz bandwidth, an average effective signal-to-noise ratio improvement of 10 dB and a clear speedup of qubit readout time., Comment: 9 pages + 2 appendix pages, 3 figures + 2 appendix figures

Anomaly action detection and localization play an essential role in security and advanced surveillance systems. However, due to the tremendous amount of surveillance videos, most of the available data for the task is unlabeled or semi-labeled with the video class known, but the location of the anomaly event is unknown. In this work, we target anomaly localization in semi-supervised videos. While the mainstream direction in addressing this task is focused on segment-level multi-instance learning and the generation of pseudo labels, we aim to explore a promising yet unfulfilled direction to solve the problem by learning the temporal relations within videos in order to locate anomaly events. To this end, we propose a hierarchical transformer model designed to evaluate the significance of observed actions in anomalous videos with a divide-and-conquer strategy along the temporal axis. Our approach segments a parent video hierarchically into multiple temporal children instances and measures the influence of the children nodes in classifying the abnormality of the parent video. Evaluating our model on two well-known anomaly detection datasets, UCF-crime and ShanghaiTech, proves its ability to interpret the observed actions within videos and localize the anomalous ones. Our proposed approach outperforms previous works relying on segment-level multiple-instance learning approaches while reaching a promising performance compared to the more recent pseudo-labeling-based approaches., Comment: Accepted at the 27th International Conference on Pattern Recognition (ICPR-2024)

Previous work shows that humans tend to prefer large bounding boxes over small bounding boxes with the same IoU. However, we show here that commonly used object detectors predict large and small boxes equally often. In this work, we investigate how to align automatically detected object boxes with human preference and study whether this improves human quality perception. We evaluate the performance of three commonly used object detectors through a user study (N = 123). We find that humans prefer object detections that are upscaled with factors of 1.5 or 2, even if the corresponding AP is close to 0. Motivated by this result, we propose an asymmetric bounding box regression loss that encourages large over small predicted bounding boxes. Our evaluation study shows that object detectors fine-tuned with the asymmetric loss are better aligned with human preference and are preferred over fixed scaling factors. A qualitative evaluation shows that human preference might be influenced by some object characteristics, like object shape., Comment: Accepted paper at the ECCV 2024 workshop on Assistive Computer Vision and Robotics (ACVR)

In recent years, computational improvements have allowed for more nuanced, data-driven and geographically explicit agent-based simulations. So far, simulations have struggled to adequately represent the attributes that motivate the actions of the agents. In fact, existing population synthesis frameworks generate agent profiles limited to socio-demographic attributes. In this paper, we introduce a novel value-enriched population synthesis framework that integrates a motivational layer with the traditional individual and household socio-demographic layers. Our research highlights the significance of extending the profile of agents in synthetic populations by incorporating data on values, ideologies, opinions and vital priorities, which motivate the agents' behaviour. This motivational layer can help us develop a more nuanced decision-making mechanism for the agents in social simulation settings. Our methodology integrates microdata and macrodata within different Bayesian network structures. This contribution allows to generate synthetic populations with integrated value systems that preserve the inherent socio-demographic distributions of the real population in any specific region.

Speech emotion recognition (SER) has made significant strides with the advent of powerful self-supervised learning (SSL) models. However, the generalization of these models to diverse languages and emotional expressions remains a challenge. We propose a large-scale benchmark to evaluate the robustness and adaptability of state-of-the-art SER models in both in-domain and out-of-domain settings. Our benchmark includes a diverse set of multilingual datasets, focusing on less commonly used corpora to assess generalization to new data. We employ logit adjustment to account for varying class distributions and establish a single dataset cluster for systematic evaluation. Surprisingly, we find that the Whisper model, primarily designed for automatic speech recognition, outperforms dedicated SSL models in cross-lingual SER. Our results highlight the need for more robust and generalizable SER models, and our benchmark serves as a valuable resource to drive future research in this direction., Comment: Accepted at INTERSPEECH 2024

Weather regimes describe the large-scale atmospheric circulation in the mid-latitudes in terms of a few circulation states that modulate regional surface weather. Subseasonal forecasts of prevailing weather regimes have proven skillful and valuable to energy applications. Previous studies have mainly focused on the mean surface weather associated with a regime. However, we show in this paper that variability of surface weather within a regime cannot be ignored. These intra-regime variations, caused by different `subflavors' of the same regime, can be captured by continuous regime indices and allow a refined application of weather regimes. Here we discuss wintertime temperature and wind speed regime anomalies for four selected countries, and provide guidance on the operational use and interpretation of regime forecasts. In an accompanying supplementary dataset we provide similar analysis for all European countries, seasons and key energy variables, useful as an applied reference., Comment: 13 pages, 4 figures, supplement in SI.pdf. Submitted to Geophysical Research Letters

Accurate localization is crucial for various applications, including autonomous vehicles and next-generation wireless networks. However, the reliability and precision of Global Navigation Satellite Systems (GNSS), such as the Global Positioning System (GPS), are compromised by multi-path errors and non-line-of-sight scenarios. This paper presents a novel approach to enhance GPS accuracy by combining visual data from RGB cameras with wireless signals captured at millimeter-wave (mmWave) and sub-terahertz (sub-THz) basestations. We propose a sensing-aided framework for (i) site-specific GPS data characterization and (ii) GPS position de-noising that utilizes multi-modal visual and wireless information. Our approach is validated in a realistic Vehicle-to-Infrastructure (V2I) scenario using a comprehensive real-world dataset, demonstrating a substantial reduction in localization error to sub-meter levels. This method represents a significant advancement in achieving precise localization, particularly beneficial for high-mobility applications in 5G and beyond networks., Comment: Datasets and code files are available on the DeepSense website: https://deepsense6g.net/. To appear in IEEE GLOBECOM 2024

It is widely accepted that cosmic rays are a plausible cause of DRAM errors in high-performance computing (HPC) systems, and various studies suggest that they could explain some aspects of the observed DRAM error behavior. However, this phenomenon is insufficiently studied in production environments. We analyze the correlations between cosmic rays and DRAM errors on two HPC clusters: a production supercomputer with server-class DDR3-1600 and a prototype with LPDDR3-1600 and no hardware error correction. Our error logs cover 2000 billion MB-hours for the MareNostrum 3 supercomputer and 135 million MB-hours for the Mont-Blanc prototype. Our analysis combines quantitative analysis, formal statistical methods and machine learning. We detect no indications that cosmic rays have any influence on the DRAM errors. To understand whether the findings are specific to systems under study, located at 100 meters above the sea level, the analysis should be repeated on other HPC clusters, especially the ones located on higher altitudes. Also, analysis can (and should) be applied to revisit and extend numerous previous studies which use cosmic rays as a hypothetical explanation for some aspects of the observed DRAM error behaviors., Comment: Accepted for publication in SBAC-PAD'24