arXiv subjects

Filippo Simini

Publications and source records attributed to Filippo Simini.

At least 19 recordsLinked to original sources

arXiv2026

Super-Resolution Initialization of High-Fidelity CFD Simulations for Pebble-Bed Reactors

High-order CFD simulations provide detailed resolution of the heterogeneous interstitial flow in pebble-bed reactors, but their computational cost is high, especially during the initial flow-development period required to reach statistically stationary conditions. This work investigates the use of a Super-Resolution Graph Neural Network (SR-GNN) to improve the initialization of high-order NekRS simulations. Lower-order P = 2 velocity fields are used as inputs to reconstruct higher-order representations, which are then used as initial conditions for P = 7 restart simulations. The approach is evaluated using a 146-pebble bed at Re = 1000, Re = 2500, and Re = 5000, with pressure-drop convergence used as the main figure of merit. The SR-GNN models were trained using paired low- and high-order snapshots and were first evaluated through qualitative inference comparisons. High-order restart simulations showed that, for Re = 1000 and Re = 2500, the SR-GNN initialized cases produced pressure-drop histories similar to direct restarts from true P = 2 fields. For Re = 5000, however, the super-resolved field restart approached the statistically stationary P = 7 pressure-drop range faster than both the direct P = 2 restart and the reference P = 7 simulation initialized from a uniform velocity field. The trained Re = 5000 model was also applied to a larger 1568-pebble bed, demonstrating qualitative applicability of the workflow to a significantly larger packed-bed geometry. These results indicate that SR-GNN-based initialization is a promising strategy for reducing high-order flow-development cost, while also motivating further work on broader Reynolds-number and geometry generalization.

physics.flu-dyn↗

arXiv2025

Sketch-Augmented Features Improve Learning Long-Range Dependencies in Graph Neural Networks

Graph Neural Networks learn on graph-structured data by iteratively aggregating local neighborhood information. While this local message passing paradigm imparts a powerful inductive bias and exploits graph sparsity, it also yields three key challenges: (i) oversquashing of long-range information, (ii) oversmoothing of node representations, and (iii) limited expressive power. In this work we inject randomized global embeddings of node features, which we term \textit{Sketched Random Features}, into standard GNNs, enabling them to efficiently capture long-range dependencies. The embeddings are unique, distance-sensitive, and topology-agnostic -- properties which we analytically and empirically show alleviate the aforementioned limitations when injected into GNNs. Experimental results on real-world graph learning tasks confirm that this strategy consistently improves performance over baseline GNNs, offering both a standalone solution and a complementary enhancement to existing techniques such as graph positional encodings. Our source code is available at \href{https://github.com/ryienh/sketched-random-features}{https://github.com/ryienh/sketched-random-features}.

cs.LG↗

arXiv2025

Aurora: Architecting Argonne's First Exascale Supercomputer for Accelerated Scientific Discovery

Aurora is Argonne National Laboratory's pioneering Exascale supercomputer, designed to accelerate scientific discovery with cutting-edge architectural innovations. Key new technologies include the Intel(TM) Xeon(TM) Data Center GPU Max Series (code-named Sapphire Rapids) with support for High Bandwidth Memory (HBM), alongside the Intel(TM) Data Center GPU Max Series (code-named Ponte Vecchio) on each compute node. Aurora also integrates the Distributed Asynchronous Object Storage (DAOS), a novel exascale storage solution, and leverages Intel's oneAPI programming environment. This paper presents an in-depth exploration of Aurora's node architecture, the HPE Slingshot interconnect, the supporting software ecosystem, and DAOS. We provide insights into standard benchmark performance and applications readiness efforts via Aurora's Early Science Program and the Exascale Computing Project.

cs.DC↗

arXiv2025

Quality Measures for Dynamic Graph Generative Models

Deep generative models have recently achieved significant success in modeling graph data, including dynamic graphs, where topology and features evolve over time. However, unlike in vision and natural language domains, evaluating generative models for dynamic graphs is challenging due to the difficulty of visualizing their output, making quantitative metrics essential. In this work, we develop a new quality metric for evaluating generative models of dynamic graphs. Current metrics for dynamic graphs typically involve discretizing the continuous-evolution of graphs into static snapshots and then applying conventional graph similarity measures. This approach has several limitations: (a) it models temporally related events as i.i.d. samples, failing to capture the non-uniform evolution of dynamic graphs; (b) it lacks a unified measure that is sensitive to both features and topology; (c) it fails to provide a scalar metric, requiring multiple metrics without clear superiority; and (d) it requires explicitly instantiating each static snapshot, leading to impractical runtime demands that hinder evaluation at scale. We propose a novel metric based on the \textit{Johnson-Lindenstrauss} lemma, applying random projections directly to dynamic graph data. This results in an expressive, scalar, and application-agnostic measure of dynamic graph similarity that overcomes the limitations of traditional methods. We also provide a comprehensive empirical evaluation of metrics for continuous-time dynamic graphs, demonstrating the effectiveness of our approach compared to existing methods. Our implementation is available at https://github.com/ryienh/jl-metric.

cs.LG↗

arXiv2024

A Deep Probabilistic Framework for Continuous Time Dynamic Graph Generation

Recent advancements in graph representation learning have shifted attention towards dynamic graphs, which exhibit evolving topologies and features over time. The increased use of such graphs creates a paramount need for generative models suitable for applications such as data augmentation, obfuscation, and anomaly detection. However, there are few generative techniques that handle continuously changing temporal graph data; existing work largely relies on augmenting static graphs with additional temporal information to model dynamic interactions between nodes. In this work, we propose a fundamentally different approach: We instead directly model interactions as a joint probability of an edge forming between two nodes at a given time. This allows us to autoregressively generate new synthetic dynamic graphs in a largely assumption free, scalable, and inductive manner. We formalize this approach as DG-Gen, a generative framework for continuous time dynamic graphs, and demonstrate its effectiveness over five datasets. Our experiments demonstrate that DG-Gen not only generates higher fidelity graphs compared to traditional methods but also significantly advances link prediction tasks.

cs.LG↗

arXiv2023

In Situ Framework for Coupling Simulation and Machine Learning with Application to CFD

Recent years have seen many successful applications of machine learning (ML) to facilitate fluid dynamic computations. As simulations grow, generating new training datasets for traditional offline learning creates I/O and storage bottlenecks. Additionally, performing inference at runtime requires non-trivial coupling of ML framework libraries with simulation codes. This work offers a solution to both limitations by simplifying this coupling and enabling in situ training and inference workflows on heterogeneous clusters. Leveraging SmartSim, the presented framework deploys a database to store data and ML models in memory, thus circumventing the file system. On the Polaris supercomputer, we demonstrate perfect scaling efficiency to the full machine size of the data transfer and inference costs thanks to a novel co-located deployment of the database. Moreover, we train an autoencoder in situ from a turbulent flow simulation, showing that the framework overhead is negligible relative to a solver time step and training epoch.

cs.LG↗

arXiv2023

A Multi-Level, Multi-Scale Visual Analytics Approach to Assessment of Multifidelity HPC Systems

The ability to monitor and interpret of hardware system events and behaviors are crucial to improving the robustness and reliability of these systems, especially in a supercomputing facility. The growing complexity and scale of these systems demand an increase in monitoring data collected at multiple fidelity levels and varying temporal resolutions. In this work, we aim to build a holistic analytical system that helps make sense of such massive data, mainly the hardware logs, job logs, and environment logs collected from disparate subsystems and components of a supercomputer system. This end-to-end log analysis system, coupled with visual analytics support, allows users to glean and promptly extract supercomputer usage and error patterns at varying temporal and spatial resolutions. We use multiresolution dynamic mode decomposition (mrDMD), a technique that depicts high-dimensional data as correlated spatial-temporal variations patterns or modes, to extract variation patterns isolated at specified frequencies. Our improvements to the mrDMD algorithm help promptly reveal useful information in the massive environment log dataset, which is then associated with the processed hardware and job log datasets using our visual analytics system. Furthermore, our system can identify the usage and error patterns filtered at user, project, and subcomponent levels. We exemplify the effectiveness of our approach with two use scenarios with the Cray XC40 supercomputer.

cs.HC↗

arXiv2022

Deep Surrogate Docking: Accelerating Automated Drug Discovery with Graph Neural Networks

The process of screening molecules for desirable properties is a key step in several applications, ranging from drug discovery to material design. During the process of drug discovery specifically, protein-ligand docking, or chemical docking, is a standard in-silico scoring technique that estimates the binding affinity of molecules with a specific protein target. Recently, however, as the number of virtual molecules available to test has rapidly grown, these classical docking algorithms have created a significant computational bottleneck. We address this problem by introducing Deep Surrogate Docking (DSD), a framework that applies deep learning-based surrogate modeling to accelerate the docking process substantially. DSD can be interpreted as a formalism of several earlier surrogate prefiltering techniques, adding novel metrics and practical training practices. Specifically, we show that graph neural networks (GNNs) can serve as fast and accurate estimators of classical docking algorithms. Additionally, we introduce FiLMv2, a novel GNN architecture which we show outperforms existing state-of-the-art GNN architectures, attaining more accurate and stable performance by allowing the model to filter out irrelevant information from data more efficiently. Through extensive experimentation and analysis, we show that the DSD workflow combined with the FiLMv2 architecture provides a 9.496x speedup in molecule screening with a <3% recall error rate on an example docking task. Our open-source code is available at https://github.com/ryienh/graph-dock.

cs.LG↗

arXiv2022

Operation-Level Performance Benchmarking of Graph Neural Networks for Scientific Applications

As Graph Neural Networks (GNNs) increase in popularity for scientific machine learning, their training and inference efficiency is becoming increasingly critical. Additionally, the deep learning field as a whole is trending towards wider and deeper networks, and ever increasing data sizes, to the point where hard hardware bottlenecks are often encountered. Emerging specialty hardware platforms provide an exciting solution to this problem. In this paper, we systematically profile and select low-level operations pertinent to GNNs for scientific computing implemented in the Pytorch Geometric software framework. These are then rigorously benchmarked on NVIDIA A100 GPUs for several various combinations of input values, including tensor sparsity. We then analyze these results for each operation. At a high level, we conclude that on NVIDIA systems: (1) confounding bottlenecks such as memory inefficiency often dominate runtime costs moreso than data sparsity alone, (2) native Pytorch operations are often as or more competitive than their Pytorch Geometric equivalents, especially at low to moderate levels of input data sparsity, and (3) many operations central to state-of-the-art GNN architectures have little to no optimization for sparsity. We hope that these results serve as a baseline for those developing these operations on specialized hardware and that our subsequent analysis helps to facilitate future software and hardware based optimizations of these operations and thus scalable GNN performance as a whole.

cs.LG↗

arXiv2020

Deep Gravity: enhancing mobility flows generation with deep neural networks and geographic information

The movements of individuals within and among cities influence critical aspects of our society, such as well-being, the spreading of epidemics, and the quality of the environment. When information about mobility flows is not available for a particular region of interest, we must rely on mathematical models to generate them. In this work, we propose the Deep Gravity model, an effective method to generate flow probabilities that exploits many variables (e.g., land use, road network, transport, food, health facilities) extracted from voluntary geographic data, and uses deep neural networks to discover non-linear relationships between those variables and mobility flows. Our experiments, conducted on mobility flows in England, Italy, and New York State, show that Deep Gravity has good geographic generalization capability, achieving a significant increase in performance (especially in densely populated regions of interest) with respect to the classic gravity model and models that do not use deep neural networks or geographic data. We also show how flows generated by Deep Gravity may be explained in terms of the geographic features using explainable AI techniques.

cs.LG↗

arXiv2020

The agglomeration and dispersion dichotomy of human settlements on Earth

Human settlements on Earth are scattered in a multitude of shapes, sizes and spatial arrangements. These patterns are often not random but a result of complex geographical, cultural, economic and historical processes that have profound human and ecological impacts. However, little is known about the global distribution of these patterns and the spatial forces that creates them. This study analyses human settlements from high-resolution satellite imagery and provides a global classification of spatial patterns. We find two emerging classes, namely agglomeration and dispersion. In the former, settlements are fewer than expected based on the predictions of scaling theory, while an unexpectedly high number of settlements characterizes the latter. Our global classification of spatial patterns correlates with some urban outcomes, such as the amount of CO2 emitted for transportation, providing insights into the relationship between land use patterns and socio-economic and environmental indicators. To explain the observed spatial patterns, we also propose a model that combines two agglomeration forces and simulates human settlements' historical growth. Our results show that our model accurately matches the observed global classification (F1: 0.73), helps to understand and estimate the growth of human settlements and, in turn, the distribution and physical dynamics of all human settlements on Earth, from small villages to cities.

physics.soc-ph↗

arXiv2019

Social media usage reveals how regions recover after natural disaster

The challenge of nowcasting and forecasting the effect of natural disasters (e.g. earthquakes, floods, hurricanes) on assets, people and society is of primary importance for assessing the ability of such systems to recover from extreme events. Traditional disaster recovery estimates, such as surveys and interviews, are usually costly, time consuming and do not scale. Here we present a methodology to indirectly estimate the post-emergency recovery status ('downtime') of small businesses in urban areas looking at their online posting activity on social media. Analysing the time series of posts before and after an event, we quantify the downtime of small businesses for three natural disasters occurred in Nepal, Puerto Rico and Mexico. A convenient and reliable method for nowcasting the post-emergency recovery status of economic activities could help local governments and decision makers to better target their interventions and distribute the available resources more effectively.

physics.soc-ph↗

arXiv2019

Scikit-mobility: a Python library for the analysis, generation and risk assessment of mobility data

The last decade has witnessed the emergence of massive mobility data sets, such as tracks generated by GPS devices, call detail records, and geo-tagged posts from social media platforms. These data sets have fostered a vast scientific production on various applications of mobility analysis, ranging from computational epidemiology to urban planning and transportation engineering. A strand of literature addresses data cleaning issues related to raw spatiotemporal trajectories, while the second line of research focuses on discovering the statistical "laws" that govern human movements. A significant effort has also been put on designing algorithms to generate synthetic trajectories able to reproduce, realistically, the laws of human mobility. Last but not least, a line of research addresses the crucial problem of privacy, proposing techniques to perform the re-identification of individuals in a database. A view on state of the art cannot avoid noticing that there is no statistical software that can support scientists and practitioners with all the aspects mentioned above of mobility data analysis. In this paper, we propose scikit-mobility, a Python library that has the ambition of providing an environment to reproduce existing research, analyze mobility data, and simulate human mobility habits. scikit-mobility is efficient and easy to use as it extends pandas, a popular Python library for data analysis. Moreover, scikit-mobility provides the user with many functionalities, from visualizing trajectories to generating synthetic data, from analyzing statistical patterns to assessing the privacy risk related to the analysis of mobility data sets.

physics.soc-ph↗

arXiv2018

Zipf's and Taylor's Laws

Zipf's law states that the frequency of an observation with a given value is inversely proportional to the square of that value; Taylor's law, instead, describes the scaling between fluctuations in the size of a population and its mean. Empirical evidence of the validity of these laws has been found in many and diverse domains. Despite the numerous models proposed to explain the presence of Zipf's law, there is no consensus on how it originates from a microscopic process of individuals dynamics without fine tuning. Here we show that Zipf's law and Taylor's law can emerge from a general class of stochastic processes at the individual level, which incorporate one of two features: environmental variability, i.e. fluctuations of parameters, or correlations, i.e. dependence between individuals. Under these assumptions, we show numerically and with theoretical arguments that the conditional variance of the population increments scales as the square of the population and that the corresponding stationary distribution of the processes follows Zipf's law.

physics.soc-ph↗

arXiv2018

Prediction of next career moves from scientific profiles

Changing institution is a scientist's key career decision, which plays an important role in education, scientific productivity, and the generation of scientific knowledge. Yet, our understanding of the factors influencing a relocation decision is very limited. In this paper we investigate how the scientific profile of a scientist determines their decision to move (i.e., change institution). To this aim, we describe a scientist's profile by three main aspects: the scientist's recent scientific career, the quality of their scientific environment and the structure of their scientific collaboration network. We then design and implement a two-stage predictive model: first, we use data mining to predict which researcher will move in the next year on the basis of their scientific profile; second we predict which institution they will choose by using a novel social-gravity model, an adaptation of the traditional gravity model of human mobility. Experiments on a massive dataset of scientific publications show that our approach performs well in both the stages, resulting in a 85% reduction of the prediction error with respect to the state-of-the-art approaches.

stat.AP↗

arXiv2017

Human Mobility: Models and Applications

Recent years have witnessed an explosion of extensive geolocated datasets related to human movement, enabling scientists to quantitatively study individual and collective mobility patterns, and to generate models that can capture and reproduce the spatiotemporal structures and regularities in human trajectories. The study of human mobility is especially important for applications such as estimating migratory flows, traffic forecasting, urban planning, and epidemic modeling. In this survey, we review the approaches developed to reproduce various mobility patterns, with the main focus on recent developments. This review can be used both as an introduction to the fundamental modeling principles of human mobility, and as a collection of technical methods applicable to specific mobility-related problems. The review organizes the subject by differentiating between individual and population mobility and also between short-range and long-range mobility. Throughout the text the description of the theory is intertwined with real-world applications.

physics.soc-ph↗

arXiv2016

Data-driven generation of spatio-temporal routines in human mobility

The generation of realistic spatio-temporal trajectories of human mobility is of fundamental importance in a wide range of applications, such as the developing of protocols for mobile ad-hoc networks or what-if analysis in urban ecosystems. Current generative algorithms fail in accurately reproducing the individuals' recurrent schedules and at the same time in accounting for the possibility that individuals may break the routine during periods of variable duration. In this article we present DITRAS (DIary-based TRAjectory Simulator), a framework to simulate the spatio-temporal patterns of human mobility. DITRAS operates in two steps: the generation of a mobility diary and the translation of the mobility diary into a mobility trajectory. We propose a data-driven algorithm which constructs a diary generator from real data, capturing the tendency of individuals to follow or break their routine. We also propose a trajectory generator based on the concept of preferential exploration and preferential return. We instantiate DITRAS with the proposed diary and trajectory generators and compare the resulting algorithm with real data and synthetic data produced by other generative algorithms, built by instantiating DITRAS with several combinations of diary and trajectory generators. We show that the proposed algorithm reproduces the statistical properties of real trajectories in the most accurate way, making a step forward the understanding of the origin of the spatio-temporal patterns of human mobility.

cs.SI↗

arXiv2016

Nanoscale artificial intelligence: creating artificial neural networks using autocatalytic reactions

A general methodology is proposed to engineer a system of interacting components (particles) which is able to self-regulate their concentrations in order to produce any prescribed output in response to a particular input. The methodology is based on the mathematical equivalence between artificial neurons in neural networks and species in autocatalytic reactions, and it specifies the relationship between the artificial neural network's parameters and the rate coefficients of the reactions between particle species. Such systems are characterised by a high degree of robustness as they are able to reach the desired output despite disturbances and perturbations of the concentrations of the various species.

nlin.AO↗