Abstract
We present the first parameterized analysis of a standard (1+1) Evolutionary Algorithm on a distribution of vertex cover problems. We show that if the planted cover is at most logarithmic, restarting the (1+1) EA every O(n log n) steps will find a cover at least as small as the planted cover in polynomial time for sufficiently dense random graphs p > 0.71. For superlogarithmic planted covers, we prove that the (1+1) EA finds a solution in fixed-parameter tractable time in expectation. We complement these theoretical investigations with a number of computational experiments that highlight the interplay between planted cover size, graph density and runtime.
Original language | English (US) |
---|---|
Title of host publication | FOGA 2023 - Proceedings of the 17th ACM/SIGEVO Conference on Foundations of Genetic Algorithms |
Publisher | Association for Computing Machinery, Inc |
Pages | 96-104 |
Number of pages | 9 |
ISBN (Electronic) | 9798400702020 |
DOIs | |
State | Published - Aug 30 2023 |
Externally published | Yes |
Event | 17th ACM/SIGEVO Conference on Foundations of Genetic Algorithms, FOGA 2023 - Potsdam, Germany Duration: Aug 30 2023 → Sep 1 2023 |
Publication series
Name | FOGA 2023 - Proceedings of the 17th ACM/SIGEVO Conference on Foundations of Genetic Algorithms |
---|
Conference
Conference | 17th ACM/SIGEVO Conference on Foundations of Genetic Algorithms, FOGA 2023 |
---|---|
Country/Territory | Germany |
City | Potsdam |
Period | 8/30/23 → 9/1/23 |
Bibliographical note
Publisher Copyright:© 2023 ACM.
Keywords
- parameterized complexity
- runtime analysis
- vertex cover