Learning Preconditions for Planning from Plan Traces and HTN Structure

Abstract

A great challenge in developing planning systems for practical applications is the difficulty of acquiring the domain information needed to guide such systems. This paper describes a way to learn some of that knowledge. More specifically: We introduce a theoretical basis for formally defining algorithms that learn preconditions for HTN methods. We describe CaMeL, a supervised, eager, and incremental learning process for preconditions of HTN methods. We state and prove theorems about CaMeL's soundness, completeness, and convergence properties. We present empirical results about CaMeL's convergence under various conditions. Among other things, CaMeL converges the fastest on the preconditions of the HTN methods that are needed the most often. Thus CaMeL's output can be useful even before it has fully converged.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2006
Accession Number
ADA448060

Entities

People

  • Dana S. Nau
  • David W. Aha
  • Hector Munoz-avila
  • Okhtay Ilghami

Organizations

  • University of Maryland

Tags

Communities of Interest

  • Autonomy
  • Counter WMD

DTIC Thesaurus Topics

  • Abstracts
  • Algorithms
  • Artificial Intelligence
  • Computer Science
  • Computers
  • Hierarchies
  • Language
  • Learning
  • Machine Learning
  • Military Operations
  • Military Research
  • Probability
  • Random Variables
  • Supervised Machine Learning
  • Test Sets
  • Training
  • Universities

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Artificial Intelligence
  • Distributed Systems and Data Platform Development