eng
Schloss Dagstuhl – Leibniz-Zentrum für Informatik
Leibniz International Proceedings in Informatics
1868-8969
2024-09-23
94:1
94:17
10.4230/LIPIcs.ESA.2024.94
article
A Row Generation Algorithm for Finding Optimal Burning Sequences of Large Graphs
Pereira, Felipe de Carvalho
1
https://orcid.org/0000-0002-8967-8576
de Rezende, Pedro Jussieu
1
https://orcid.org/0000-0002-9529-4253
Yunes, Tallys
2
https://orcid.org/0000-0002-8308-7812
Morato, Luiz Fernando Batista
1
https://orcid.org/0000-0003-0938-2941
Institute of Computing, University of Campinas, Brazil
Miami Herbert Business School, University of Miami, Coral Gables, FL, USA
We propose an exact algorithm for the Graph Burning Problem (GBP), an NP-hard optimization problem that models the spread of influence on social networks. Given a graph G with vertex set V, the objective is to find a sequence of k vertices in V, namely, v₁, v₂, … , v_k, such that k is minimum and ⋃_{i=1}^{k} {u∈V: d(u,v_i) ≤ k-i} = V, where d(u,v) denotes the distance between u and v. We formulate the problem as a set covering integer programming model and design a row generation algorithm for the GBP. Our method exploits the fact that a very small number of covering constraints is often sufficient for solving the integer model, allowing the corresponding rows to be generated on demand. To date, the most efficient exact algorithm for the GBP, denoted here by GDCA, is able to obtain optimal solutions for graphs with up to 14,000 vertices within two hours of execution. In comparison, our algorithm finds provably optimal solutions approximately 236 times faster, on average, than GDCA. For larger graphs, memory space becomes a limiting factor for GDCA. Our algorithm, however, solves real-world instances with more than 3 million vertices in less than 19 minutes, increasing the size of graphs for which optimal solutions are known by a factor of 200. Additionally, we conduct tests on the proposed algorithm using a series of challenging instances composed of grid graphs containing up to 5,000 vertices. As a result, we achieve novel optimal solutions and tight optimality gaps that have not been previously reported in the literature.
https://drops.dagstuhl.de/storage/00lipics/lipics-vol308-esa2024/LIPIcs.ESA.2024.94/LIPIcs.ESA.2024.94.pdf
Graph Burning
Burning Number
Burning Sequence
Set Covering
Integer Programming
Row Generation