A Neuro-Symbolic Approach for Probabilistic Reasoning on Graph Data
Raffaele Pojer, Andrea Passerini, Kim G. Larsen, Manfred Jaeger · 2025
Read A Neuro-Symbolic Approach for Probabilistic Reasoning on Graph Data by Raffaele Pojer, Andrea Passerini, Kim G. Larsen, Manfred Jaeger (2025) 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-2507-21873.txt.
Description
Raffaele Pojer and three co-authors merge neural networks with symbolic probability models in a 2025 paper on graph data. Graph neural networks predict well on linked data, but they cannot use stated domain knowledge and they answer only narrow prediction questions. An older family, relational Bayesian networks, builds full probability models over graph-like data and answers many kinds of questions, but learns less sharply from data.
Their framework aims to keep both strengths. The neural side supplies fast pattern learning, while the symbolic side supplies stated rules and broad reasoning about chance. That pairing could open graph tasks where users need not just a label but a reasoned probability, such as risk checks that must respect known constraints. The paper lays out the approach and its building blocks.
Catalog record
| CREATOR | Raffaele Pojer, Andrea Passerini, Kim G. Larsen, Manfred Jaeger |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 70 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-21873