LipnLab
🎉 Two papers from the #MachineLearning and #NLP teams @LipnLab were accepted to #ICML!
▶️ The paper "Delaunay Graph: Addressing Over-Squashing and Over-Smoothing Using Delaunay Triangulation" by H. Attali, D. Buscladi, N. Pernelle presents a novel graph rewiring method that incorporates node features with low complexity to alleviate both Over-Squashing and Over-Smoothing issues.
🔗 https://sites.google.com/view/hugoattali/research?authuser=0
Hugo Attali - Research

My research interests lie in how to quantify the role of graph topology in GNNs and to what extent we can improve the structural properties of the input graph to better exchange messages between layers.

May 13, 2024 at 3:10pmWeb
Show threadLipnLabMay 13, 2024
▶️ The paper "Predicting Lagrangian Multipliers for Mixed Integer Linear Programs" by F. Demelas, J. Le Roux, M. Lacroix, A. Parmentier (CERMICS)"
introduces a deep learning methodology to predict Lagrangian multipliers for problems with compact formulation, where Lagrangian Relaxation provides better bounds compared to Continuous Relaxation.