This paper presents Inc*, a general algorithm that can be used in conjunction with any local search heuristic and that has the potential to substantially improve the overall performance of the heuristic. Genetic programming is used to discover new strategies for the Inc* algorithm. We experimentally compare performance of local heuristics for SAT with and without the Inc* algorithm. Results show that Inc* consistently improves performance.
|Name||Lecture notes in computer science|
|Conference||Proceedings of the 8th European Conference, Evolutionary Computation in Combinatorial Optimization|
|Period||26/03/08 → 28/03/08|
- Genetic Programming
- Boolean Satisfiability Problems (SAT)