Can route planning be smarter with transfer optimization?
Abstract
We aim to showcase the benefit of transfer optimization for route planning problems by illustrating how the solution accuracy of travelling salesman problem instances can be enhanced via autonomous and positive transfer of knowledge from related source problems that have been encountered previously. Our approach is able to achieve better solution accuracy by exploiting useful past experiences at runtime, based on a source-target similarity measure learned online.
Authors: Ray Lim, Yew-Soon Ong, Hanh Thi Hong Phan, Abhishek Gupta, Allan Neng Sheng Zhang
Published in: Genetic and Evolutionary Computation Conference Companion (GECCO Companion) (2019)