Speaker: Yvan Dérick KAPTCHOUANG
Abstract
This talk addresses the challenge of energy efficiency in Graph Neural Networks (GNNs), whose growing complexity leads to high computational and energy costs. We present a generic methodology for designing frugal Graph Convolutional Networks (GCNs) based on the combination of parallel graph processing on multicore architectures and lightweight compression techniques. The proposed approach significantly reduces execution time and energy consumption while preserving high predictive performance, and is applicable across a wide range of graph-structured datasets, including biomedical, citation, and large-scale real-world graphs
Speaker Bio
Yvan Dérick KAPTCHOUANG holds a Master’s degree in Computer Science with a specialization in Data Science from the University of Yaoundé I. Their research focuses on frugal machine learning, particularly the use of parallelization and compression techniques to improve the energy efficiency and scalability of graph-based neural networks (GNNs). Yvan has experience in NLP, computer vision, and document analysis, and has authored international research submissions. Their broader interests include Green AI and high-performance computing for efficient deep learning.