Public articles linked to the same research event.
arXiv This survey examines rule-based languages for neurosymbolic AI, including Datalog, answer set programs, and probabilistic logic programs, along four axes: semantics, expressiveness, neural integration, and evaluation mechanism; it analyses over 50 recent systems and applications and compares formalism usage across databases and programming languages, machine learning, vision, and robotics, provides a decision matrix mapping six application scenarios to required features, concludes that no single dialect covers every scenario while provenance semirings come closest to a common foundation, and outlines five open problems.
This survey examines rule-based languages for neurosymbolic AI, including Datalog, answer set programs, and probabilistic logic programs, along four axes: semantics, expressiveness, neural integration, and evaluation mechanism; it analyses over 50 recent systems and applications and compares formalism usage across databases and programming languages, machine learning, vision, and robotics, provides a decision matrix mapping six application scenarios to required features, concludes that no single dialect covers every scenario while provenance semirings come closest to a common foundation, and outlines five open problems.
This survey examines rule-based languages for neurosymbolic AI, including Datalog, answer set programs, and probabilistic logic programs, along four axes: semantics, expressiveness, neural integration, and evaluation mechanism; it analyses over 50 recent systems and applications and compares formalism usage across databases and programming languages, machine learning, vision, and robotics, provides a decision matrix mapping six application scenarios to required features, concludes that no single dialect covers every scenario while provenance semirings come closest to a common foundation, and outlines five open problems.
This survey examines rule-based languages for neurosymbolic AI, including Datalog, answer set programs, and probabilistic logic programs, along four axes: semantics, expressiveness, neural integration, and evaluation mechanism; it analyses over 50 recent systems and applications and compares formalism usage across databases and programming languages, machine learning, vision, and robotics, provides a decision matrix mapping six application scenarios to required features, concludes that no single dialect covers every scenario while provenance semirings come closest to a common foundation, and outlines five open problems.