Neurosymbolische Integration
Wir entwickeln neurosymbolische Integration sowohl auf konzeptioneller Ebene als auch in verschiedenen Anwendungsbereichen, darunter Chemie, Materialwissenschaft und menschliches Verhalten. Wir nutzen gut entwickeltes ontologisches Hintergrundwissen, um die Leistung von maschinellen Lernmethoden zu verbessern, z.B. durch die Verwendung einer semantischen Verlustfunktion oder von Neuro-Fuzzy-Regeln. Wir haben eine neuartige neurosymbolische Architektur für die ontologische Klassifikation von strukturierten Entitäten und für die Erweiterung von Ontologien entwickelt.
Laufende Projekte
- StrOntEx (Ontology Extension by Automated Learning and Reasoning from Structured Entities) funded by DFG
Abgeschlossene Projekte
Veröffentlichungen
2025
Chebifier 2: An Ensemble for Chemistry.
In: S. Tiwari, editor, Symbolic and Generative AI for Science (SymGenAI4Sci 2025). 2025. To appear
Simon Flügel, Martin Glauer, Janna Hastings, Till Mossakowski, Christopher J. Mungall, Charlotte Tumescheit and Fabian Neuhaus.
[BibTeX]
ChemLog: Making MSOL Viable for Ontological Classification and Learning.
In: A. Tamaddoni-Nezhad, editor, 5th International Joint Conference on Learning and Reasoning (IJCLR 2025) . 2025. To appear
Simon Flügel, Martin Glauer, Till Mossakowski and Fabian Neuhaus.
[BibTeX]
Box Embeddings for Extending Ontologies: A Data-Driven and Interpretable Approach.
Journal of Cheminformatics, 17(138), 2025.
Adel Memariani, Martin Glauer, Simon Flügel, Fabian Neuhaus, Janna Hastings and Till Mossakowski.
[doi] [BibTeX]
Monadic ULLER: A Unified Categorical Semantics of the Neurosymbolic ULLER Framework.
In: G. Eleonora, P. Hitzler and E. van Krieken, editors, 19th Conference on Neurosymbolic Learning and Reasoning (NeSy 2025). 2025. to appear
Daniel Schellhorn and Till Mossakowski.
[BibTeX]
Advancing Natural Language formalization to First Order Logic with Fine-tuned LLMs.
In: A. Tamaddoni-Nezhad, editor, 5th International Joint Conference on Learning and Reasoning (IJCLR 2025). 2025. To appear
Felix Vossel, Till Mossakowski and Björn Gehrke.
[BibTeX]
2024
A fuzzy loss for ontology classification.
In: T. R. Besold, A. d'Avila Garcez, E. Jimenez-Ruiz, R. Confalonieri, B. Wagner and P. Madhyastha, editors, NeSy 2024: The 18th International Conference on Neural-symbolic Learning and Reasoning, volume 14979, series Springer lecture notes, pages 101-118. Springer, 2024.
Simon Flügel, Martin Glauer, Till Mossakowski and Fabian Neuhaus.
[abstract] [BibTeX]
Chebifier: Automating semantic classification in ChEBI to accelerate data-driven discovery.
Digital Discovery, 2024.
Martin Glauer, Fabian Neuhaus, Simon Flügel, Marie Wosny, Till Mossakowski, Adel Memariani, Johannes Schwerdt and Janna Hastings.
[doi] [abstract] [BibTeX]
Interpretable Ontology Extension in Chemistry.
Semantic Web journal, 15(4):937-958, 2024. Special Issue on The Role of Ontologies and Knowledge in Explainable AI
Martin Glauer, Adel Memariani, Fabian Neuhaus, Till Mossakowski and Janna Hastings.
[doi] [BibTeX]
2023
Neuro-symbolic semantic learning for chemistry.
In: P. Hitzler, M. K. Sarker and A. Eberhart, editors, A Compendium of Neuro-Symbolic Artificial Intelligence, chapter 21, pages 460 - 484. IOS press, 2023.
Martin Glauer, Till Mossakowski, Fabian Neuhaus, Adel Memariani and Janna Hastings.
[doi] [abstract] [BibTeX]
Ontology Pre-training for Poison Prediction.
In: D. Seipel and A. Steen, editors, German conference on artificial intelligence 2023, volume 14236, series Lecture Notes in Artificial Intelligence, pages 31-45. Springer, 2023. Best paper award. Also available at doi.org/10.48550/arXiv.2301.08577
Martin Glauer, Fabian Neuhaus, Till Mossakowski and Janna Hastings.
[doi] [abstract] [BibTeX]
2022
Modular design patterns for neural-symbolic integration: refinement and combination.
In: A. d'Avila Garcez and E. Jiménez-Ruiz, editors, 16th International Workshop on Neural-Symbolic Learning and Reasoning (NeSy), volume 3212, series CEUR Workshop proceedings, pages 192-201. 2022.
Till Mossakowski.
[doi] [BibTeX]
2021
Learning chemistry: exploring the suitability of machine learning for the task of structure-based chemical ontology classification..
Journal of Cheminformatics, 13(23), 2021.
Janna Hastings, Martin Glauer, Adel Memariani, Fabian Neuhaus and Till Mossakowski.
[doi] [BibTeX]
Automated and explainable ontology extension based on deep learning: A case study in the chemical domain.
In: R. Confalonieri, O. Kutz and D. Calvanese, editors, International Workshop on Data meets Applied Ontologies in Explainable AI (DAO-XAI 2021), volume 2998, series CEUR Workshop Proceedings. ceur-ws.org/Vol-2998/, 2021.
Adel Memariani, Martin Glauer, Fabian Neuhaus, Till Mossakowski and Janna Hastings.
[doi] [BibTeX]
2019
Towards Fuzzy Neural Conceptors.
IfCoLog Journal of Logics and their Applications, 6(4):725-744, 2019.
Till Mossakowski, Razvan Diaconescu and Martin Glauer.
[doi] [BibTeX]