Graph Neural Networks and Arithmetic Circuits

Authors: Timon Barlag, Vivian Holzapfel, Laura Strieker, Jonni Virtema, Heribert Vollmer

NeurIPS 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical 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.
Researcher Affiliation Academia Timon Barlag Institute for Theoretical Computer Science Leibniz University Hanover Hanover, Germany Vivian Holzapfel Institute for Theoretical Computer Science Leibniz University Hanover Hanover, Germany Laura Strieker Institute for Theoretical Computer Science Leibniz University Hanover Hanover, Germany Jonni Virtema School of Computer Science University of Sheffield Sheffield, United Kingdom Heribert Vollmer Institute for Theoretical Computer Science Leibniz University Hanover Hanover, Germany
Pseudocode No The paper contains figures illustrating circuit examples, but no structured pseudocode or algorithm blocks are provided.
Open Source Code No The paper is purely theoretical and does not mention releasing any source code for its methodology. The NeurIPS checklist indicates 'NA' for questions regarding code and data.
Open Datasets No The paper is theoretical and does not involve datasets or training. The NeurIPS checklist indicates 'NA' for questions regarding experimental reproducibility.
Dataset Splits No The paper is theoretical and does not involve dataset splits for validation. The NeurIPS checklist indicates 'NA' for questions regarding experimental reproducibility.
Hardware Specification No The paper is theoretical and does not conduct experiments, thus no hardware specifications are mentioned. The NeurIPS checklist indicates 'NA' for questions regarding experimental resources.
Software Dependencies No The paper is theoretical and does not describe any software dependencies with specific version numbers. The NeurIPS checklist indicates 'NA' for questions regarding experimental resources.
Experiment Setup No The paper is theoretical and does not conduct experiments, thus no experimental setup details like hyperparameters are provided. The NeurIPS checklist indicates 'NA' for questions regarding experimental settings.