Learning Hierarchical Policies by Iteratively Reducing the Width of Sketch Rules.

Dominik Drexler, Jendrik Seipp, Hector Geffner

Browse the full KR paper archive.