Low-Rank Projections of GCNs Laplacian - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2021

Low-Rank Projections of GCNs Laplacian

Nathan Grinsztajn
  • Function : Author
  • PersonId : 1083320
Philippe Preux
Edouard Oyallon

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.
Fichier principal
Vignette du fichier
main.pdf (534.12 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03248056 , version 1 (03-06-2021)

Identifiers

Cite

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

Altmetric

Share

Gmail Facebook X LinkedIn More