PhD Proposal: Curriculum Learning for Hierarchical Task Networks

Hierarchical Task Networks (HTNs) require a domain expert to provide knowledgeabout the dynamics of the planning domain, including structural properties andpotential hierarchical problem-solving strategies. A significant knowledge engineeringburden for a domain expert is required to write HTN decomposition methods. Sometechniques overcome this burden, in part, by learning HTN methods after analyzingthe semantics of a solution trace for a sample planning problem.My proposed research will focus on improving HTN learning using curricula. Acurriculum may improve an HTN learner by teaching it to learn simpler methodsbefore learning gradually more complex ones that make use of previously learnedmethods. Such a curriculum may be generated from landmarks, each of which is afact that is true at some point in every plan that solves the problem.In this dissertation proposal, first, I describe the completed algorithm HTNMakerC,a modified version of HTN-Maker that learns HTN methods from acurriculum, along with the theoretical analysis and discussion of the experimentresults. I propose to develop an algorithm that uses landmarks to generate curriculathat include interesting tasks to learn in each sample planning problem. Furthermore,a more general curriculum learning technique would learn HTN methods from differentplanning problems and their solution traces, and those problems are organized in away that the HTN methods learned from the previous problems could be utilized inlearning HTN methods from the later problems. Therefore, I propose to develop analgorithm that performs curriculum learning among a sequence of sample problems.In addition, I propose an algorithm that learns unlimitedly recursive HTN methodsthat may increase the problem coverage of the set of learned methods. I also describemy plan for the theoretical analysis of the proposed algorithms, their implementation,and their evaluation in a variety of domains and problem sets from the InternationalPlanning Competitions.

Examining Committee

Chair:

Dr. Dana Nau

Department Representative:

Dr. Dave Levin

Members:

Dr. Pratap Tokekar

Dr. Mark Roberts (NRL)