Pascal Bouvry
YOU?
Author Swipe
View article: Portable <scp>PGAS</scp> ‐Based <scp>GPU</scp> ‐Accelerated Branch‐And‐Bound Algorithms at Scale
Portable <span>PGAS</span> ‐Based <span>GPU</span> ‐Accelerated Branch‐And‐Bound Algorithms at Scale Open
The Branch‐and‐Bound (B&B) technique plays a key role in solving many combinatorial optimization problems, enabling efficient problem‐solving and decision‐making in a wide range of applications. It incrementally constructs a tree by buildi…
View article: A Review on Quantum Circuit Optimization using ZX-Calculus
A Review on Quantum Circuit Optimization using ZX-Calculus Open
Quantum computing promises significant speed-ups for certain algorithms but the practical use of current noisy intermediate-scale quantum (NISQ) era computers remains limited by resources constraints (e.g., noise, qubits, gates, and circui…
View article: Trustworthy Efficient Communication for Distributed Learning using LQ-SGD Algorithm
Trustworthy Efficient Communication for Distributed Learning using LQ-SGD Algorithm Open
We propose LQ-SGD (Low-Rank Quantized Stochastic Gradient Descent), an efficient communication gradient compression algorithm designed for distributed training. LQ-SGD further develops on the basis of PowerSGD by incorporating the low-rank…
View article: Lightweight Trustworthy Distributed Clustering
Lightweight Trustworthy Distributed Clustering Open
Ensuring data trustworthiness within individual edge nodes while facilitating collaborative data processing poses a critical challenge in edge computing systems (ECS), particularly in resource-constrained scenarios such as autonomous syste…
View article: Exhaustive Search for Quantum Circuit Optimization using ZX Calculus
Exhaustive Search for Quantum Circuit Optimization using ZX Calculus Open
Quantum computers allow a near-exponential speed-up for specific applications when compared to classical computers. Despite recent advances in the hardware of quantum computers, their practical usage is still severely limited due to a rest…
View article: Multi-objective methods in Federated Learning: A survey and taxonomy
Multi-objective methods in Federated Learning: A survey and taxonomy Open
The Federated Learning paradigm facilitates effective distributed machine learning in settings where training data is decentralized across multiple clients. As the popularity of the strategy grows, increasingly complex real-world problems …
View article: GPU-accelerated Tree Search in Chapel
GPU-accelerated Tree Search in Chapel Open
This repository contains the implementation of a GPU-accelerated tree search algorithm in Chapel. The latter allows to solve instances of the N-Queens and permutation flowshop scheduling problems. For comparison purpose, CUDA-based counter…
View article: A Unified Hyperparameter Optimization Pipeline for Transformer-Based Time Series Forecasting Models
A Unified Hyperparameter Optimization Pipeline for Transformer-Based Time Series Forecasting Models Open
Transformer-based models for time series forecasting (TSF) have attracted significant attention in recent years due to their effectiveness and versatility. However, these models often require extensive hyperparameter optimization (HPO) to …
View article: Training Green AI Models Using Elite Samples
Training Green AI Models Using Elite Samples Open
The substantial increase in AI model training has considerable environmental implications, requiring energy-efficient and sustainable AI practices. On one hand, data-centric approaches show great potential towards training energy-efficient…
View article: FedPref: Federated Learning Across Heterogeneous Multi-objective Preferences
FedPref: Federated Learning Across Heterogeneous Multi-objective Preferences Open
The Federated Learning paradigm is a distributed machine learning strategy, developed for settings where training data is owned by distributed devices and cannot be shared with others. Federated Learning circumvents this constraint by carr…
View article: A Reproducible Study and Performance Analysis of GPU Programming Paradigms: OpenACC vs. CUDA in Key Linear Algebra Computations
A Reproducible Study and Performance Analysis of GPU Programming Paradigms: OpenACC vs. CUDA in Key Linear Algebra Computations Open
Scientific and engineering problems are frequently governed by partial differential equations; however, the analytical solutions of these equations are often impractical, thereby forcing the adoption of numerical methods. Basic Linear Alge…
View article: Productivity- and Performance-aware Parallel Distributed Depth-First Search
Productivity- and Performance-aware Parallel Distributed Depth-First Search Open
Chapel-based Productivity- and Performance-aware Parallel Distributed Depth-First Search algorithm, named P3D-DFS. The latter is a generic algorithm that can be instantiated on numerous tree-based problems. It is based on the DistBag-DFS d…
View article: Trustworthiness of Stochastic Gradient Descent in Distributed Learning
Trustworthiness of Stochastic Gradient Descent in Distributed Learning Open
Distributed learning (DL) uses multiple nodes to accelerate training, enabling efficient optimization of large-scale models. Stochastic Gradient Descent (SGD), a key optimization algorithm, plays a central role in this process. However, co…
View article: Dataset | Mindset = Explainable AI | Interpretable AI
Dataset | Mindset = Explainable AI | Interpretable AI Open
We often use "explainable" Artificial Intelligence (XAI)" and "interpretable AI (IAI)" interchangeably when we apply various XAI tools for a given dataset to explain the reasons that underpin machine learning (ML) outputs. However, these n…
View article: Heterogeneity: An Open Challenge for Federated On-board Machine Learning
Heterogeneity: An Open Challenge for Federated On-board Machine Learning Open
The design of satellite missions is currently undergoing a paradigm shift from the historical approach of individualised monolithic satellites towards distributed mission configurations, consisting of multiple small satellites. With a rapi…
View article: Survey and Taxonomy: The Role of Data-Centric AI in Transformer-Based Time Series Forecasting
Survey and Taxonomy: The Role of Data-Centric AI in Transformer-Based Time Series Forecasting Open
Alongside the continuous process of improving AI performance through the development of more sophisticated models, researchers have also focused their attention to the emerging concept of data-centric AI, which emphasizes the important rol…
View article: GPU-Accelerated Tree-Search in Chapel Versus CUDA and HIP
GPU-Accelerated Tree-Search in Chapel Versus CUDA and HIP Open
peer reviewed
View article: Round-Based Mechanism and Job Packing with Model-Similarity-Based Policy for Scheduling DL Training in GPU Cluster
Round-Based Mechanism and Job Packing with Model-Similarity-Based Policy for Scheduling DL Training in GPU Cluster Open
Graphics Processing Units (GPUs) are employed for their parallel processing capabilities, which are essential to train deep learning (DL) models with large datasets within a reasonable time. However, the diverse GPU architectures exhibit v…
View article: Training Green AI Models Using Elite Samples
Training Green AI Models Using Elite Samples Open
The substantial increase in AI model training has considerable environmental implications, mandating more energy-efficient and sustainable AI practices. On the one hand, data-centric approaches show great potential towards training energy-…
View article: An Edge-Based Approach to Partitioning and Overlapping Graph Clustering with User-Specified Density
An Edge-Based Approach to Partitioning and Overlapping Graph Clustering with User-Specified Density Open
Graph clustering has received considerable attention recently, and its applications are numerous, ranging from the detection of social communities to the clustering of computer networks. It is classified as an NP-class problem, and several…
View article: Transformer Multivariate Forecasting: Less is More?
Transformer Multivariate Forecasting: Less is More? Open
In the domain of multivariate forecasting, transformer models stand out as powerful apparatus, displaying exceptional capabilities in handling messy datasets from real-world contexts. However, the inherent complexity of these datasets, cha…
View article: Towards Unified Data Ingestion and Transfer for the Computing Continuum
Towards Unified Data Ingestion and Transfer for the Computing Continuum Open
peer reviewed
View article: Constraint Model for the Satellite Image Mosaic Selection Problem
Constraint Model for the Satellite Image Mosaic Selection Problem Open
Satellite imagery solutions are widely used to study and monitor different regions of the Earth. However, a single satellite image can cover only a limited area. In cases where a larger area of interest is studied, several images must be s…