Exploring continual learning strategies in artificial neural networks through graph-based analysis of connectivity: Insights from a brain-inspired perspective.

Journal: Neural networks : the official journal of the International Neural Network Society
PMID:

Abstract

Artificial Neural Networks (ANNs) aim at mimicking information processing in biological networks. In cognitive neuroscience, graph modeling is a powerful framework widely used to study brain structural and functional connectivity. Yet, the extension of graph modeling to ANNs has been poorly explored especially in terms of functional connectivity (i.e. the contextual change of the activity's units in networks). In the perspective of designing more robust and interpretable ANNs, we study how a brain-inspired graph-based approach can be extended and used to investigate ANN properties and behaviors. We focus our study on different continual learning strategies inspired by the biological mechanisms and modeled with ANNs. We show that graph modeling offers a simple and elegant framework to deeply investigate ANNs, compare their performances, and explore deleterious behaviors such as catastrophic forgetting.

Authors

  • Lucrezia Carboni
    Univ. Grenoble Alpes, Inria, CNRS, LJK, Grenoble, France; Univ. Grenoble Alpes, Inserm, U1216, Grenoble Institut Neurosciences, GIN, Grenoble, France.
  • Dwight Nwaigwe
    Univ. Grenoble Alpes, Inria, CNRS, LJK, Grenoble, France; Univ. Grenoble Alpes, Inserm, U1216, Grenoble Institut Neurosciences, GIN, Grenoble, France.
  • Marion Mainsant
    Univ. Grenoble Alpes, Univ. Savoie Mont Blanc, CNRS, LPNC, 38000 Grenoble, France; Univ. Grenoble Alpes, CEA, LIST, 38000 Grenoble, France.
  • Raphael Bayle
    Univ. Grenoble Alpes, CEA, LIST, 38000 Grenoble, France.
  • Marina Reyboz
    Univ. Grenoble Alpes, CEA, LIST, 38000 Grenoble, France.
  • Martial Mermillod
    University Grenoble Alpes, University Savoie Mont Blanc, CNRS, LPNC, 38000 Grenoble, France; University Grenoble Alpes, CNRS, Grenoble INP, LJK, 38000 Grenoble, France. Electronic address: Martial.Mermillod@univ-grenoble-alpes.fr.
  • Michel Dojat
    Univ. Grenoble Alpes, Inserm, U1216, Grenoble Institut Neurosciences, GIN, Grenoble, France. Electronic address: michel.dojat@inserm.fr.
  • Sophie Achard
    Univ. Grenoble Alpes, Inria, CNRS, LJK, Grenoble, France.