{"slug": "from-neural-networks-to-logical-theories", "title": "From Neural Networks to Logical Theories", "summary": "A paper submitted on 28 Sep 2025 to arXiv's Computer Science > Artificial Intelligence section formalizes the correspondence between fibring of modal logics and fibring of neural networks, a gap the authors say was never formally established. Using fibred models compatible with fibred neural networks, the paper derives non-uniform logical expressiveness results for Graph Neural Networks (GNNs), Graph Attention Networks (GATs) and Transformer encoders. The authors state the longer-term goal is to open the way for fibring as a formalism for interpreting the logical theories learnt by neural networks with the tools of computational logic.", "body_md": "# Computer Science > Artificial Intelligence\n\n  [Submitted on 28 Sep 2025]\n\n# Title:From Neural Networks to Logical Theories: The Correspondence between Fibring Modal Logics and Fibring Neural Networks\n\n[View PDF](/pdf/2509.23912)\n\n[HTML (experimental)](https://arxiv.org/html/2509.23912v1)\n\nAbstract:Fibring of modal logics is a well-established formalism for combining countable families of modal logics into a single fibred language with common semantics, characterized by fibred models. Inspired by this formalism, fibring of neural networks was introduced as a neurosymbolic framework for combining learning and reasoning in neural networks. Fibring of neural networks uses the (pre-)activations of a trained network to evaluate a fibring function computing the weights of another network whose outputs are injected back into the original network. However, the exact correspondence between fibring of neural networks and fibring of modal logics was never formally established. In this paper, we close this gap by formalizing the idea of fibred models \\emph{compatible} with fibred neural networks. Using this correspondence, we then derive non-uniform logical expressiveness results for Graph Neural Networks (GNNs), Graph Attention Networks (GATs) and Transformer encoders. Longer-term, the goal of this paper is to open the way for the use of fibring as a formalism for interpreting the logical theories learnt by neural networks with the tools of computational logic.\n    \n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer \n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers \n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps \n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations \n\n*(*[What are Smart Citations?](https://www.scite.ai/))\n# Code, Data and Media Associated with this Article\n\nalphaXiv \n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers \n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub \n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub \n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face \n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast \n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))\n# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower \n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender \n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))\n# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [**Learn more about arXivLabs**](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/from-neural-networks-to-logical-theories", "canonical_source": "https://arxiv.org/abs/2509.23912", "published_at": "2026-09-12 18:04:53+00:00", "updated_at": "2026-09-12 18:21:41.914558+00:00", "lang": "en", "topics": ["artificial-intelligence", "neural-networks", "ai-research", "machine-learning", "natural-language-processing"], "entities": ["arXiv", "Graph Neural Networks", "Graph Attention Networks", "Transformer"], "alternates": {"html": "https://wpnews.pro/news/from-neural-networks-to-logical-theories", "markdown": "https://wpnews.pro/news/from-neural-networks-to-logical-theories.md", "text": "https://wpnews.pro/news/from-neural-networks-to-logical-theories.txt", "jsonld": "https://wpnews.pro/news/from-neural-networks-to-logical-theories.jsonld"}}