Loading [MathJax]/extensions/tex2jax.js

Graph Neural Networks and Arithmetic Circuits

Publikation: Beitrag in Fachzeitschrift › Konferenzaufsatz in Fachzeitschrift › Forschung › Peer-Review

Autorschaft

Organisationseinheiten

Externe Organisationen

  • The University of Sheffield

Details

OriginalspracheEnglisch
FachzeitschriftAdvances in Neural Information Processing Systems
Jahrgang37
PublikationsstatusVeröffentlicht - 2024
Veranstaltung38th Conference on Neural Information Processing Systems, NeurIPS 2024 - Vancouver, Kanada
Dauer: 10 Dez. 2024 → 15 Dez. 2024

Abstract

We characterize the computational power of neural networks that follow the graph neural network (GNN) architecture, not restricted to aggregate-combine GNNs or other particular types. We establish an exact correspondence between the expressivity of GNNs using diverse activation functions and arithmetic circuits over real numbers. In our results the activation function of the network becomes a gate type in the circuit. Our result holds for families of constant depth circuits and networks, both uniformly and non-uniformly, for all common activation functions.

ASJC Scopus Sachgebiete

Zitieren

Graph Neural Networks and Arithmetic Circuits. / Barlag, Timon; Holzapfel, Vivian; Strieker, Laura et al.
in: Advances in Neural Information Processing Systems, Jahrgang 37, 2024.

Publikation: Beitrag in Fachzeitschrift › Konferenzaufsatz in Fachzeitschrift › Forschung › Peer-Review

Barlag, T, Holzapfel, V, Strieker, L, Virtema, J & Vollmer, H 2024, 'Graph Neural Networks and Arithmetic Circuits', Advances in Neural Information Processing Systems, Jg. 37. https://doi.org/10.48550/arXiv.2402.17805
Barlag, T., Holzapfel, V., Strieker, L., Virtema, J., & Vollmer, H. (2024). Graph Neural Networks and Arithmetic Circuits. Advances in Neural Information Processing Systems, 37. https://doi.org/10.48550/arXiv.2402.17805
Barlag T, Holzapfel V, Strieker L, Virtema J, Vollmer H. Graph Neural Networks and Arithmetic Circuits. Advances in Neural Information Processing Systems. 2024;37. doi: 10.48550/arXiv.2402.17805
Barlag, Timon ; Holzapfel, Vivian ; Strieker, Laura et al. / Graph Neural Networks and Arithmetic Circuits. in: Advances in Neural Information Processing Systems. 2024 ; Jahrgang 37.
Download
@article{15477634a8a54c62bad9e954f94fb5fb,
title = "Graph Neural Networks and Arithmetic Circuits",
abstract = "We characterize the computational power of neural networks that follow the graph neural network (GNN) architecture, not restricted to aggregate-combine GNNs or other particular types. We establish an exact correspondence between the expressivity of GNNs using diverse activation functions and arithmetic circuits over real numbers. In our results the activation function of the network becomes a gate type in the circuit. Our result holds for families of constant depth circuits and networks, both uniformly and non-uniformly, for all common activation functions.",
author = "Timon Barlag and Vivian Holzapfel and Laura Strieker and Jonni Virtema and Heribert Vollmer",
note = "Publisher Copyright: {\textcopyright} 2024 Neural information processing systems foundation. All rights reserved.; 38th Conference on Neural Information Processing Systems, NeurIPS 2024 ; Conference date: 10-12-2024 Through 15-12-2024",
year = "2024",
doi = "10.48550/arXiv.2402.17805",
language = "English",
volume = "37",

}

Download

TY - JOUR

T1 - Graph Neural Networks and Arithmetic Circuits

AU - Barlag, Timon

AU - Holzapfel, Vivian

AU - Strieker, Laura

AU - Virtema, Jonni

AU - Vollmer, Heribert

N1 - Publisher Copyright: © 2024 Neural information processing systems foundation. All rights reserved.

PY - 2024

Y1 - 2024

N2 - We characterize the computational power of neural networks that follow the graph neural network (GNN) architecture, not restricted to aggregate-combine GNNs or other particular types. We establish an exact correspondence between the expressivity of GNNs using diverse activation functions and arithmetic circuits over real numbers. In our results the activation function of the network becomes a gate type in the circuit. Our result holds for families of constant depth circuits and networks, both uniformly and non-uniformly, for all common activation functions.

AB - We characterize the computational power of neural networks that follow the graph neural network (GNN) architecture, not restricted to aggregate-combine GNNs or other particular types. We establish an exact correspondence between the expressivity of GNNs using diverse activation functions and arithmetic circuits over real numbers. In our results the activation function of the network becomes a gate type in the circuit. Our result holds for families of constant depth circuits and networks, both uniformly and non-uniformly, for all common activation functions.

UR - http://www.scopus.com/inward/record.url?scp=105000491402&partnerID=8YFLogxK

U2 - 10.48550/arXiv.2402.17805

DO - 10.48550/arXiv.2402.17805

M3 - Conference article

AN - SCOPUS:105000491402

VL - 37

JO - Advances in Neural Information Processing Systems

JF - Advances in Neural Information Processing Systems

SN - 1049-5258

T2 - 38th Conference on Neural Information Processing Systems, NeurIPS 2024

Y2 - 10 December 2024 through 15 December 2024

ER -

Von denselben Autoren