Workshop on Foundations of Neuro-Symbolic Logical Reasoning

As part of TriCoLore 2026

Held in Brixen – Bressanone, Italy in early December.

Neurosymbolic AI is seen as a way to increase trustworthiness and interpretability of subsymbolic learning approaches by incorporating structured information, such as ontologies,  and symbolic reasoning.

Neurosymbolic AI can, however, not only be seen from a subsymbolic perspective, thus to enhance subsymbolic approaches by symbolic reasoning but also the other way round.

In particular, logics, especially non-classical logics suffer from issues regarding representational and computational complexity and struggle to be applied to real-world use cases.

Here, neurosymbolic approaches can come in handy to allow for an enhancement of logical approaches with subsymbolic techniques, e.g., by symbol grounding.

This allows for widening the appeal of complex logical frameworks by simpler access through a NeSy framework.

This workshop aims at bringing together researchers from the neurosymbolic AI community with logicians and ontologists to explore the synergy between these fields.