Approximation of Binary-Valued Functions by Networks of Finite VC Dimension

0577075 - ÚI 2024 RIV CH eng C - Conference Paper (international conference)
Kůrková, Věra
Approximation of Binary-Valued Functions by Networks of Finite VC Dimension.
Artificial Neural Networks and Machine Learning – ICANN 2023. Proceedings, Part I. Cham: Springer, 2023 - (Iliadis, L.; Papaleonidas, A.; Angelov, P.; Jayne, C.), s. 483-490. Lecture Notes in Computer Science, 14254. ISBN 978-3-031-44206-3. ISSN 0302-9743.
[ICANN 2023: International Conference on Artificial Neural Networks /32./. Heraklion (GR), 26.09.2023-29.09.2023]
R&D Projects: GA ČR(CZ) GA22-02067S
Institutional support: RVO:67985807
Keywords : approximation by neural networks * bounds on approximation errors * VC dimension * growth function * high-dimensional probability * concentration inequalities * method of bounded differences
OECD category: Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
https://dx.doi.org/10.1007/978-3-031-44207-0_40

Distributions of errors in approximation of binary-valued functions by networks with sets of input-output functions of finite VC dimension is investigated. Conditions on concentration of approximation errors around their mean values are derived in terms of growth functions of sets of input-output functions. Limitations of approximation capabilities of networks of finite VC dimension are discussed.
Permanent Link: https://hdl.handle.net/11104/0346341