QUESTIONS.md

Questions we have in the development process

1. What is a good score function to FPs?
2. What is neighborhood relation for FPs?
3. Should SLS work well on kenken examples?
4. What is a good score function for FP predicates like fp.normal?
5. How to reduce the number of neighbors? (e.g., take range into consideration)

General Questions

1. Why not use simulated annealing?
2. Can we use FP->BV and link it with Z3 SLS?