Skip to Main content Skip to Navigation
New interface
Conference papers

Low-Rank Projections of GCNs Laplacian

Nathan Grinsztajn 1 Philippe Preux 1 Edouard Oyallon 2 
1 Scool - Scool
Inria Lille - Nord Europe, CRIStAL - Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189
Abstract : In this work, we study the behavior of standard models for community detection under spectral manipulations. Through various ablation experiments, we evaluate the impact of bandpass filtering on the performance of a GCN: we empirically show that most of the necessary and used information for nodes classification is contained in the low-frequency domain, and thus contrary to images, high frequencies are less crucial to community detection. In particular, it is sometimes possible to obtain accuracies at a state-of-the-art level with simple classifiers that rely only on a few low frequencies.
Complete list of metadata
Contributor : Edouard Oyallon Connect in order to contact the contributor
Submitted on : Thursday, June 3, 2021 - 12:41:26 PM
Last modification on : Tuesday, November 22, 2022 - 2:26:16 PM
Long-term archiving on: : Saturday, September 4, 2021 - 6:36:28 PM


Files produced by the author(s)


  • HAL Id : hal-03248056, version 1
  • ARXIV : 2106.07360


Nathan Grinsztajn, Philippe Preux, Edouard Oyallon. Low-Rank Projections of GCNs Laplacian. ICLR 2021 Workshop GTRL, May 2021, Online, France. ⟨hal-03248056⟩



Record views


Files downloads