Progress has been made recently in developing techniques to automatically generate effective heuristics.These techniques typically aim to reduce the size of the search tree, usually by combining more primitive heuristics. However, simply reducing search tree size is not enough to guarantee that problems will be solved more quickly. We describe a new approach to automatic heuristic generation that combines more primitive heuristics in a way that can produce better heuristics than current methods. We report on experiments using 14 planning domains that show our system leads to a much greater reduction in search time than previous methods. In closing, we discuss avenues for extending this promising approach to combining heuristics.
|Title of host publication||Proceedings of the Twenty-Fourth International Conference on Automated Planning and Scheduling|
|Editors||Steve Chien, Alan Fern, Wheeler Ruml|
|Number of pages||9|
|Publication status||Published - 10 May 2014|
|Event||24th International Conference on Automated Planning and Scheduling - Portsmouth, United States|
Duration: 21 Jun 2014 → 26 Jun 2014
Conference number: 24
http://icaps14.icaps-conference.org/ (Link to Conference Website)
|Conference||24th International Conference on Automated Planning and Scheduling|
|Abbreviated title||ICAPS 2014|
|Period||21/06/14 → 26/06/14|
Barley, M. W., Franco, S., & Riddle, P. J. (2014). Overcoming the Utility Problem in Heuristic Generation: Why Time Matters. In S. Chien, A. Fern, & W. Ruml (Eds.), Proceedings of the Twenty-Fourth International Conference on Automated Planning and Scheduling (pp. 38-46). AAAI press.