The Differentiable Feasibility Pump
Cacciola, Matteo, Forel, Alexandre, Frangioni, Antonio, Lodi, Andrea
–arXiv.org Artificial Intelligence
Although nearly 20 years have passed since its conception, the feasibility pump algorithm remains a widely used heuristic to find feasible primal solutions to mixed-integer linear problems. Many extensions of the initial algorithm have been proposed. Yet, its core algorithm remains centered around two key steps: solving the linear relaxation of the original problem to obtain a solution that respects the constraints, and rounding it to obtain an integer solution. This paper shows that the traditional feasibility pump and many of its follow-ups can be seen as gradient-descent algorithms with specific parameters. A central aspect of this reinterpretation is observing that the traditional algorithm differentiates the solution of the linear relaxation with respect to its cost. This reinterpretation opens many opportunities for improving the performance of the original algorithm. We study how to modify the gradient-update step as well as extending its loss function. We perform extensive experiments on MIPLIB instances and show that these modifications can substantially reduce the number of iterations needed to find a solution.
arXiv.org Artificial Intelligence
Nov-5-2024
- Country:
- Asia > Middle East
- Republic of Türkiye > Istanbul Province > Istanbul (0.04)
- Europe
- Italy > Tuscany
- Pisa Province > Pisa (0.04)
- Middle East > Republic of Türkiye
- Istanbul Province > Istanbul (0.04)
- United Kingdom > England
- Cambridgeshire > Cambridge (0.04)
- Italy > Tuscany
- North America
- Canada > Quebec
- Montreal (0.04)
- United States > New York
- New York County > New York City (0.04)
- Canada > Quebec
- Asia > Middle East
- Genre:
- Research Report > New Finding (0.46)
- Technology: