Search arXiv⌕ Search

arXiv subjects

Fuad Hasan

Publications and source records attributed to Fuad Hasan.

8 recordsLinked to original sources
arXiv2026

Keep Your Friends Close, and the Right Neighbours Closer: Disaster-Conditioned Kernel-Regularized Graph Attention for Building Damage Classification

Disaster damage is spatial: buildings rarely fail in isolation. Yet using spatial context for damage classification remains surprisingly underexplored, and many pipelines still rely primarily on per-building appearance cues even when the dominant uncertainty is spatially structured. Complicating matters, the right neighbourhood is not the same across events. Floods, hurricanes, and wildfires can exhibit very different clustering behaviour, making spatial reasoning valuable but easy to misuse - naive context aggregation can improve visual coherence while oversmoothing boundaries or propagating structured errors. We study this tension on xBD (the dataset used in the xView2 challenge) in a controlled post-localization, classification-only setup: each building is represented by a pre/post combined (PPC) patch cropped from the provided polygons, and spatial context is modelled with GPS-derived building graphs. Our approach keeps local evidence "close" by preserving strong spatial relationships in disaster damage patterns, while bringing only the right neighbours "closer" through a disaster-type-conditioned graph model that injects a learnable multi-scale spatial kernel prior into attention, allowing the effective neighbourhood scale to adapt across disaster types rather than being learned as a single global smoothing rule. To discourage coherence-by-smoothing, we add a residual de-correlation loss that penalizes positive Moran's~I in prediction residuals. We evaluate the method under event and dataset shift with a leave-one-event-out (LOEO) protocol on xBD and cross-dataset transfer from xBD to Ida-BD. The model improves macro-F1 and substantially reduces residual spatial autocorrelation under zero-shot event shift, indicating better use of spatial context rather than naive smoothing and enabling more reliable transfer to unseen events within known disaster types.

cs.CV↗
arXiv2026

Unstructured Mesh Tools for Fusion Energy System Design

The execution of accurate simulations of fusion energy systems requires the appropriate representation of critical component geometries as well as the coupling of complex fusion physics codes with one another and with engineering analysis tools. This paper examines the challenges of creating simulation workflows that fully leverage existing fusion research codes while integrating them with commercial computer-aided engineering (CAE) software. Key areas addressed include: (a) the construction and meshing of analysis geometries taking full advantage of available geometric modeling and meshing technologies; (b) the effective coupling of fusion physics and engineering analysis codes; and (c) the support for simulation workflows that couple particle and continuum modeling methods.

cs.CE↗
arXiv2025

PCMS: Parallel Coupler For Multimodel Simulations

This paper presents the Parallel Coupler for Multimodel Simulations (PCMS), a new GPU accelerated generalized coupling framework for coupling simulation codes on leadership class supercomputers. PCMS includes distributed control and field mapping methods for up to five dimensions. For field mapping PCMS can utilize discretization and field information to accommodate physics constraints. PCMS is demonstrated with a coupling of the gyrokinetic microturbulence code XGC with a Monte Carlo neutral transport code DEGAS2 and with a 5D distribution function coupling of an energetic particle transport code (GNET) to a gyrokinetic microturbulence code (GTC). Weak scaling is also demonstrated on up to 2,080 GPUs of Frontier with a weak scaling efficiency of 85%.

cs.DC↗
arXiv2025

GPU Acceleration of Monte Carlo Tallies on Unstructured Meshes in OpenMC with PUMI-Tally

Unstructured mesh tallies are a bottleneck in Monte Carlo neutral particle transport simulations of fusion reactors. This paper introduces the PUMI-Tally library that takes advantage of mesh adjacency information to accelerate these tallies on CPUs and GPUs. For a fixed source simulation using track-length tallies, we achieved a speed-up of 19.7X on an NVIDIA A100, and 9.2X using OpenMP on 128 threads of two AMD EPYC 7763 CPUs on NERSC Perlmutter. On the Empire AI alpha system, we achieved a speed-up of 20X using an NVIDIA H100 and 96 threads of an Intel Xenon 8568Y+. Our method showed better scaling with number of particles and number of elements. Additionally, we observed a 199X reduction in the number of allocations during initialization and the first three iterations, with a similar overall memory consumption. And, our hybrid CPU/GPU method demonstrated a 6.69X improvement in the energy consumption over the current approach.

cs.DC↗
arXiv2023

An End-to-End Vehicle Trajcetory Prediction Framework

Anticipating the motion of neighboring vehicles is crucial for autonomous driving, especially on congested highways where even slight motion variations can result in catastrophic collisions. An accurate prediction of a future trajectory does not just rely on the previous trajectory, but also, more importantly, a simulation of the complex interactions between other vehicles nearby. Most state-of-the-art networks built to tackle the problem assume readily available past trajectory points, hence lacking a full end-to-end pipeline with direct video-to-output mechanism. In this article, we thus propose a novel end-to-end architecture that takes raw video inputs and outputs future trajectory predictions. It first extracts and tracks the 3D location of the nearby vehicles via multi-head attention-based regression networks as well as non-linear optimization. This provides the past trajectory points which then feeds into the trajectory prediction algorithm consisting of an attention-based LSTM encoder-decoder architecture, which allows it to model the complicated interdependence between the vehicles and make an accurate prediction of the future trajectory points of the surrounding vehicles. The proposed model is evaluated on the large-scale BLVD dataset, and has also been implemented on CARLA. The experimental results demonstrate that our approach outperforms various state-of-the-art models.

cs.CV↗
arXiv2021

Shock Induced Damage Mechanism Of Perineuronal Net

ECM components, such as the Perineuronal net (PNN), one of the most prevalent parts surrounding the neuronal cell. PNN is a protective net-like structure regulating neuronal activity such as neurotransmission, charge balance and generates an action potential. Shock induced damage of this essential component may cause neuronal cell death and potentially leads to CTE, AD diseases, PTSD, etc. The shock generated possibly during an accident, improvised devie explosion or collision between NFL players may lead to damage to this safety net. The goal is to investigate the mechanics of PNN under shock wave. To understand the mechanics of PNN, mechanical properties of different PNN components such as glycan, GAG, and protein need to be evaluated. In this study, we evaluated the mechanical strength of PNN molecules and the interfacial strength between the components of PNN. Afterward, we have assessed the PNN molecules' damage efficiency at various conditions such as shock speed, preexisting bubble, and boundary conditions. The secondary structure altercation of the protein molecules of the PNN has been analyzed to evaluate damage intensity under varying shock loading. At higher shock speed, damage intensity is more elevated, and hyaluronan is most likely to break at the rigid junction. The primary structure of the protein molecules is most unlikely to fail. Instead, the molecules' secondary bonds will be altered. Our study suggests that the number of hydrogen bonds during the shock wave propagation decreased.

physics.bio-ph↗
arXiv2020

Effect of Random Fiber Network and Fracture Toughness on the Onset of Cavitation in Soft Materials

Experimental and theoretical observations have agreed that the onset of cavitation in soft materials requires higher tensile pressure than pure water. The extra tensile pressure is required since the cavitating bubble needs to overcome the elastic energy in soft materials. In this manuscript, we have developed two models to study and quantify the extra tensile pressure. In the first approach, we proposed a strain energy based random fiber network (RFN) failure criteria in which interaction between the cavitating bubble and RFN is considered. Gelatin samples are prepared for different concentrations, and SEM images are used to study the microstructural properties of the RFN. A unit-cell model is introduced to evaluate the geometrical and mechanical properties of the RFN. The network strain energy formulation is then coupled with the bubble growth, and the critical condition is set as the fibers ultimate failure strain. We considered soft materials as homogeneous hyper-elastic Ogden material, and fracture-based failure criteria are proposed in the second approach. The critical energy release rate is considered for quantifying the extra tensile pressure. Both the models are then compared with the existing cavitation onset criteria of rubber like materials. The validation is done with the experimental results of threshold tensile pressure for different gelatin concentrations. We have found that due to the large distribution of the pore size in the network, the nucleation pressure is similar to water. Both models can moderately predict the extra tensile pressure within the intermediate range of gelatin concentrations. For low concentration, the network's non-affinity plays a significant role and must be incorporated. On the other hand, for higher concentrations, the entropic deformation dominates, and strain energy formulation is not adequate.

cond-mat.soft↗