Abstract
The large-step Markov chain (LSMC) approach is the most effective known heuristic for large symmetric TSP instances; cf. recent results of [Martin, Otto and Felten, 1991] and [Johnson, 1990]. In this paper, we examine relationships among (i) the underlying local optimization engine within the LSMC approach, (ii) the “kick move” perturbation that is applied between successive local search descents, and (iii) the resulting LSMC solution quality. We find that the traditional “double-bridge” kick move is not necessarily optimum: stronger local optimization engines (e.g., Lin-Kernighan) are best matched with stronger kick moves. We also propose use of an adaptive temperature schedule to allow escape from deep basins of attraction; the resulting hierarchical LSMC variant outperforms traditional LSMC implementations that use uniformly zero temperatures. Finally, a population-based LSMC variant is studied, wherein multiple solution paths can interact to achieve improved solution quality.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.