Efficient Solutions for Probabilistic Planning

• Grant number: 13/11724-0
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: October 01, 2013
• End date: June 13, 2014
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Leliane Nunes de Barros
• Grantee: Felipe Werndl Trevizan
• Host Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

Probabilistic planning is a general framework to represent decision problems in which actions have probabilistic effects. Due to the importance of such framework, it has being studied in several different areas, e.g., operations research, control theory, economics and robotics. Different from other areas, artificial intelligence has focused on domain-independent probabilistic planning, that is, techniques that do not assume anything about the problem being solved and therefore can be directly applied to any probabilistic planning problem.In this project, we propose to improve domain-independent probabilistic planners in both performance (speed) and scalability. Due to the domain-independent assumption, the obtained improvements will impact not only the artificial intelligence community, but also other areas of knowledge that use the framework of probabilistic planning, e.g., public health, ecology, sustainability, economics and robotics. Our proposed approach consists in both extending the state-of-the-art algorithms and techniques, including the short-sighted probabilistic planning framework introduced during the proponent's PhD, and developing new algorithms and heuristics for domain-independent probabilistic planning.

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

TREVIZAN, FELIPE W.; VELOSO, MANUELA M.. Depth-based short-sighted stochastic shortest path problems. ARTIFICIAL INTELLIGENCE, v. 216, p. 179-205, NOV 2014. (13/11724-0)