Answer Set Programming

Answer Set Programming represents problems as rules with defaults and exceptions, then uses stable-model semantics to find solutions that satisfy the rules and constraints. Systems typically ground programs and use solver search; weak constraints support optimization, while extensions such as ASP modulo theories and temporal ASP handle numeric and time-based problems. Applications include scheduling, trajectory planning, compliance, and visual question answering, with newer neuro-symbolic approaches combining ASP with neural models or using language models to generate rules that are checked by solvers. A central challenge is scaling grounding and search while integrating learned components without losing the clarity and reliability of declarative semantics.