The Logical Expressiveness of Topological Neural Networks

Amirreza Akbari, Amauri H. Souza, Vikas Garg · 2026

Read The Logical Expressiveness of Topological Neural Networks by Amirreza Akbari, Amauri H. Souza, Vikas Garg (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-19212.txt.

Description

Akbari, Souza, and Garg ask how logically expressive topological neural networks are. Graph neural networks lead learning on graphs, yet their power is limited, often framed through the Weisfeiler-Leman hierarchy or first-order logic. Topological networks add higher-order relational structure to message passing, promising more representational power. The paper pins down which binary classifiers such networks can represent, by analyzing isomorphism tests drawn from their inner mechanisms.

Catalog record

CREATORAmirreza Akbari, Amauri H. Souza, Vikas Garg
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE144538
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.19212

Browse the full library · About the Tower of Babel