TOPOLOGICALLY-CONSISTENT MAGNITUDE PRUNING FOR VERY LIGHTWEIGHT GRAPH CONVOLUTIONAL NETWORKS - Archive ouverte HAL Access content directly
Conference Papers Year : 2022

TOPOLOGICALLY-CONSISTENT MAGNITUDE PRUNING FOR VERY LIGHTWEIGHT GRAPH CONVOLUTIONAL NETWORKS

Hichem Sahbi
  • Function : Author
  • PersonId : 1181870

Abstract

Graph convolution networks (GCNs) are currently mainstream in learning with irregular data. These models rely on message passing and attention mechanisms that capture context and node-to-node relationships. With multi-head attention, GCNs become highly accurate but oversized, and their deployment on cheap devices requires their pruning. However, pruning at high regimes usually leads to topologically inconsistent networks with weak generalization. In this paper, we devise a novel method for lightweight GCN design. Our proposed approach parses and selects subnetworks with the highest magnitudes while guaranteeing their topological consistency. The latter is obtained by selecting only accessible and co-accessible connections which actually contribute in the evaluation of the selected subnetworks. Experiments conducted on the challenging FPHA dataset show the substantial gain of our topologically consistent pruning method especially at very high pruning regimes.
Fichier principal
Vignette du fichier
article.pdf (306.19 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03838830 , version 1 (03-11-2022)

Identifiers

Cite

Hichem Sahbi. TOPOLOGICALLY-CONSISTENT MAGNITUDE PRUNING FOR VERY LIGHTWEIGHT GRAPH CONVOLUTIONAL NETWORKS. IEEE International Conference on Image Processing (ICIP), Oct 2022, Bordeaux, France. pp.3495-3499, ⟨10.1109/ICIP46576.2022.9897899⟩. ⟨hal-03838830⟩
12 View
163 Download

Altmetric

Share

Gmail Facebook X LinkedIn More