{"type":"ScholarlyArticle","title":"NeSyA: Neurosymbolic Automata","authors":[{"name":"Nikolaos Manginas","slug":"nikolaos-manginas","given":"Nikolaos","family":"Manginas","affiliation":""},{"name":"George Paliouras","slug":"george-paliouras","given":"George","family":"Paliouras","affiliation":""},{"name":"Luc De Raedt","slug":"luc-de-raedt","given":"Luc","family":"de Raedt","affiliation":""}],"abstract":"Neurosymbolic (NeSy) AI has emerged as a promising direction to integrate\r\nneural and symbolic reasoning. Unfortunately, little effort has been given \r\nto developing NeSy systems tailored to sequential/temporal problems. We identify\r\nsymbolic automata (which combine the power of automata for temporal reasoning\r\nwith that of propositional logic for static reasoning) as a suitable formalism for \r\nexpressing knowledge in temporal domains. Focusing on the task of sequence classification\r\nand tagging we show that symbolic automata can be integrated with neural-based\r\nperception, under probabilistic semantics towards an end-to-end differentiable model. \r\nOur proposed hybrid model, termed NeSyA (Neuro Symbolic \r\nAutomata) is shown to either scale or perform more accurately than previous\r\nNeSy systems in a synthetic benchmark and to provide benefits in terms of generalization\r\ncompared to purely neural systems in a real-world event recognition task.\r\nCode is available at: https://github.com/nmanginas/nesya","doi":"10.24963/ijcai.2025/662","url":"https://ijcai.org/proceedings/2025/662","pdf":"https://ijcai.org/proceedings/2025/0662.pdf","markdown":"https://ijcai.org/proceedings/2025/662.md","bibtex":"https://ijcai.org/proceedings/2025/bibtex/662","volume":{"slug":"2025","title":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2025","url":"https://ijcai.org/proceedings/2025/","year":2025,"location":"Montreal, Canada","dates":"16-22 August 2025","editors":"James Kwok","isbn":"978-1-956792-06-5","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2025"},"section":"Machine Learning","track":"Main Track","number":662,"pages":{"first":5950,"last":5958},"keywords":["Machine Learning: ML: Neuro-symbolic methods/Abductive Learning","Knowledge Representation and Reasoning: KRR: Learning and reasoning"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}