Learning to rank for synthesizing planning heuristics

We investigate learning heuristics for domainspecific planning. Prior work framed learning a heuristic as an ordinary regression problem. However, in a greedy best-first search, the ordering of states induced by a heuristic is more indicative of the resulting planner’s performance than mean squared error. Thus, we instead frame learning a heuristic as a learning to rank problem which we solve using a RankSVM formulation. Additionally, we introduce new methods for computing features that capture temporal interactions in an approximate plan. Our experiments on recent International Planning Competition problems show that the RankSVM learned heuristics outperform both the original heuristics and heuristics learned through ordinary regression.National Science Foundation (U.S.) (Grant 1420927)National Science Foundation (U.S.) (Grant 1420927)National Science Foundation (U.S.) (Grant 1523767)United States. Office of Naval Research (Grant N00014-14-1-0486)United States. Army Research Office (Grant W911NF1410433