What question did this study set out to answer?

This work aims to investigate logical frameworks as query languages for machine learning models, especially neural networks.

February 27, 2026Open Access

Query Languages for Machine-Learning Models (Invited Talk)

Key Points

This work aims to investigate logical frameworks as query languages for machine learning models, especially neural networks.
Discussion of first-order logic with summation and its recursive extension.
Illustrative examples showcasing queries expressible in these logics.
Analysis of fundamental results related to expressiveness and computational complexity.
Established connections between query capabilities and the structure of neural networks.
Demonstrated how the discussed logics enhance the understanding of model expressiveness.
Highlighted significant implications for computational complexity in the context of neural network queries.

Abstract

In my invited talk and this accompanying paper, I discuss two logics for weighted finite structures: first-order logic with summation (FO(SUM)) and its recursive extension IFP(SUM). These logics originate from foundational work by Grädel, Gurevich, and Meer in the 1990s. In recent joint work with Standke, Steegmans, and Van den Bussche, we have investigated these logics as query languages for machine learning models, specifically neural networks, which are naturally represented as weighted graphs. I present illustrative examples of queries to neural networks that can be expressed in these logics and discuss fundamental results on their expressiveness and computational complexity.

اسأل الذكاء الاصطناعي

Bookmark

View Full Paper

Cite This Study

Martin Grohe (Thu,) studied this question.

synapsesocial.com/papers/69a1357fed1d949a99abf6d8 https://doi.org/https://doi.org/10.4230/lipics.stacs.2026.1

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

اسأل الذكاء الاصطناعي

Bookmark

View Full Paper