Learning domain-independent planning heuristics with hypergraph networks

William Shen, Felipe Trevizan, Sylvie Thiébaux

    Research output: Contribution to journalConference articlepeer-review

    38 Citations (Scopus)


    We present the first approach capable of learning domain-independent planning heuristics entirely from scratch. The heuristics we learn map the hypergraph representation of the delete-relaxation of the planning problem at hand, to a cost estimate that approximates that of the least-cost path from the current state to the goal through the hypergraph. We generalise Graph Networks to obtain a new framework for learning over hypergraphs, which we specialise to learn planning heuristics by training over state/value pairs obtained from optimal cost plans. Our experiments show that the resulting architecture, STRIPS-HGNS, is capable of learning heuristics that are competitive with existing delete-relaxation heuristics including LM-cut. We show that the heuristics we learn are able to generalise across different problems and domains, including to domains that were not seen during training.

    Original languageEnglish
    Pages (from-to)574-584
    Number of pages11
    JournalProceedings International Conference on Automated Planning and Scheduling, ICAPS
    Publication statusPublished - 29 May 2020
    Event30th International Conference on Automated Planning and Scheduling, ICAPS 2020 - Nancy, France
    Duration: 26 Oct 202030 Oct 2020

    Cite this