Skip to Main content Skip to Navigation
Book sections

Graphs as Tools to Improve Deep Learning Methods

Carlos Lassance 1 Myriam Bontonou 2, 3 Mounia Hamidouche 4 Bastien Pasdeloup 5, 4 Lucas Drumetz 2, 6 Vincent Gripon 2, 3
3 Lab-STICC_2AI - Equipe Algorithm Architecture Interactions
Lab-STICC - Laboratoire des sciences et techniques de l'information, de la communication et de la connaissance : UMR6285
4 Lab-STICC_DECIDE - Equipe DECIDE
Lab-STICC - Laboratoire des sciences et techniques de l'information, de la communication et de la connaissance : UMR6285
6 Lab-STICC_OSE - Equipe Observations Signal & Environnement
Lab-STICC - Laboratoire des sciences et techniques de l'information, de la communication et de la connaissance : UMR6285
Abstract : In recent years, deep neural networks (DNNs) have known an important rise in popularity. However, although they are state-of-the-art in many machine learning challenges, they still suffer from several limitations. For example, DNNs require a lot of training data, which might not be available in some practical applications. In addition, when small perturbations are added to the inputs, DNNs are prone to misclassification errors. DNNs are also viewed as black-boxes and as such their decisions are often criticized for their lack of interpretability. In this chapter, we review recent works that aim at using graphs as tools to improve deep learning methods. These graphs are defined considering a specific layer in a deep learning architecture. Their vertices represent distinct samples, and their edges depend on the similarity of the corresponding intermediate representations. These graphs can then be leveraged using various methodologies, many of which built on top of graph signal processing. This chapter is composed of four main parts: tools for visualizing intermediate layers in a DNN, denoising data representations, optimizing graph objective functions and regularizing the learning process.
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03377563
Contributor : Myriam Bontonou Connect in order to contact the contributor
Submitted on : Thursday, October 14, 2021 - 11:17:11 AM
Last modification on : Thursday, October 28, 2021 - 4:07:05 AM
Long-term archiving on: : Saturday, January 15, 2022 - 6:34:40 PM

File

2110.03999.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03377563, version 1
  • ARXIV : 2110.03999

Citation

Carlos Lassance, Myriam Bontonou, Mounia Hamidouche, Bastien Pasdeloup, Lucas Drumetz, et al.. Graphs as Tools to Improve Deep Learning Methods. Advances in Artificial Intelligence: Reviews, 2022. ⟨hal-03377563⟩

Share

Metrics

Les métriques sont temporairement indisponibles