Constraint Programming
Also known as CP
Overview
Constraint programming represents a problem with variables, domains, and constraints, then searches for assignments that satisfy the rules or optimize an objective. Solvers combine propagation, which prunes impossible values, with branching, global constraints for structures such as scheduling and routing, and methods such as SAT/ILP integration or logic-based Benders decomposition. Applications include resource and workforce scheduling, routing, timetabling, and configuration, where complex rules make flexible models useful. Current work explores stronger propagation, hybrid and learned search, and AI-assisted modeling; a central challenge is improving performance while keeping models and results correct and verifiable.
Latest papers
40 papers3 papers in the last four weeks, level with the four weeks before. 0.0% of all new papers.
Inner-Learning: From Hand-Coded Beliefs to Declarative Knowledge in Trick-Taking Games
Solving Flood-It Puzzles: A Comparison of Tailored A* Search with Declarative Solver Approaches
Learning Size Independent Constraint Models from Input Data and Positive and Negative Example Solutions
Using Weisfeiler-Leman Features for Algorithm Selection in Constraint Optimisation
WLc) designed to model structural partitions and provide a more nuanced predictive signal. We evaluate our approach on instances from the 2023--2025 MiniZinc Challenges across two tasks: maximizing Borda count scores and maximizing predictive accuracy. Experimental results across Support Vector Machines, Random Forests, and Multi-Layer Perceptrons demonstrate that cut-based features outperform fzn2feat with SVMs, while results with RFs and MLPs are closer.