Abstract | ||
---|---|---|
Markov Decision Processes are a powerful framework for planning under uncertainty, but current algorithms have difficulties scaling to large problems. We present a novel probabilistic planner based on the notion of hybridizing two algorithms. In particular, we hybridize GPT, an exact MDP solver, with MBP, a planner that plans using a qualitative (nondeterministic) model of uncertainty. Whereas exact MDP solvers produce optimal solutions, qualitative planners sacrifice optimality to achieve speed and high scalability. Our hybridized planner, HYBPLAN, is able to obtain the best of both techniques -- speed, quality and scalability. Moreover, HYBPLAN has excellent anytime properties and makes effective use of available time and memory. |
Year | Venue | Keywords |
---|---|---|
2007 | IJCAI | markov decision processes,hybridized planner,stochastic domain,current algorithm,novel probabilistic planner,qualitative planners sacrifice optimality,exact mdp solvers,high scalability,effective use,available time,exact mdp solver,markov decision process |
Field | DocType | Citations |
Mathematical optimization,Nondeterministic algorithm,Computer science,Markov decision process,Planner,Artificial intelligence,Solver,Probabilistic logic,Scaling,Machine learning,Scalability | Conference | 7 |
PageRank | References | Authors |
0.98 | 19 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mausam | 1 | 1571 | 87.91 |
Piergiorgio Bertoli | 2 | 775 | 46.89 |
Daniel S. Weld | 3 | 10298 | 1127.49 |