Convolutional Neural Networks as 2-D systems
Optimization and Control
2023-04-12 v2 Machine Learning
Systems and Control
Systems and Control
Abstract
This paper introduces a novel representation of convolutional Neural Networks (CNNs) in terms of 2-D dynamical systems. To this end, the usual description of convolutional layers with convolution kernels, i.e., the impulse responses of linear filters, is realized in state space as a linear time-invariant 2-D system. The overall convolutional Neural Network composed of convolutional layers and nonlinear activation functions is then viewed as a 2-D version of a Lur'e system, i.e., a linear dynamical system interconnected with static nonlinear components. One benefit of this 2-D Lur'e system perspective on CNNs is that we can use robust control theory much more efficiently for Lipschitz constant estimation than previously possible.
Keywords
Cite
@article{arxiv.2303.03042,
title = {Convolutional Neural Networks as 2-D systems},
author = {Dennis Gramlich and Patricia Pauli and Carsten W. Scherer and Frank Allgöwer and Christian Ebenbauer},
journal= {arXiv preprint arXiv:2303.03042},
year = {2023}
}