Training morphological neural networks with gradient descent: some theoretical insights

Samy Blusseau (1)

1 CMM - Centre de Morphologie Mathématique

Morphological neural networks, or layers, can be a powerful tool to boost the progress in mathematical morphology, either on theoretical aspects such as the representation of complete lattice operators, or in the development of image processing pipelines. However, these architectures turn out to be difficult to train when they count more than a few morphological layers, at least within popular machine learning frameworks which use gradient descent based optimization algorithms. In this paper we investigate the potential and limitations of differentiation based approaches and back-propagation applied to morphological networks, in light of the non-smooth optimization concept of Bouligand derivative. We provide insights and first theoretical guidelines, in particular regarding initialization and learning rates.

Mots clés: Morphological neural networks Nonsmooth optimization Lattice operators; Morphological neural networks; Nonsmooth optimization; Lattice operators

Domaines: Machine Learning [stat.ML]

Samy Blusseau. Training morphological neural networks with gradient descent: some theoretical insights. IAPR Third International Conference on Discrete Geometry and Mathematical Morphology, Andrea Frosini; Elena Barcucci; Elisa Pergola; Michela Ascolese; Niccoló Di Marco; Simone Rinaldi; Sara Brunetti; Giulia Palma; Veronica Gierrini; Leonardo Bindi, Apr 2024, Firenze, Italy. ⟨hal-04438062⟩