EPFL | Biomedical Imaging Group | Lipschitz Functions

Approximation of Lipschitz Functions Using Deep Spline Neural Networks

S. Neumayer, A. Goujon, P. Bohra, M. Unser

SIAM Journal on Mathematics of Data Science, vol. 5, no. 2, pp. 306-322, 2023.

Although Lipschitz-constrained neural networks have many applications in machine learning, the design and training of expressive Lipschitz-constrained networks is very challenging. Since the popular rectified linear-unit networks have provable disadvantages in this setting, we propose using learnable spline activation functions with at least three linear regions instead. We prove that our choice is universal among all componentwise 1-Lipschitz activation functions in the sense that no other weight-constrained architecture can approximate a larger class of functions. Additionally, our choice is at least as expressive as the recently introduced non-componentwise Groupsort activation function for spectral-norm-constrained weights. The theoretical findings of this paper are consistent with previously published numerical results.

@ARTICLE(http://bigwww.epfl.ch/publications/neumayer2301.html,
AUTHOR="Neumayer, S. and Goujon, A. and Bohra, P. and Unser, M.",
TITLE="Approximation of {L}ipschitz Functions Using Deep Spline Neural
	Networks",
JOURNAL="{SIAM} Journal on Mathematics of Data Science",
YEAR="2023",
volume="5",
number="2",
pages="306--322",
month="",
note="")

© 2023 SIAM. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from SIAM. This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.