Exploiting Belief Locality in Run-Time Decision-Theoretic Planners


While Partially-Observable Markov Decision Processes have become a popular means of representing realistic planning problems, exact approaches to finding POMDP policies are extremely computationally complex. An alternative approach for control in POMDP domains is to use run-time optimization over action sequences in a dynamic decision network. While exact… (More)

2 Figures and Tables


  • Presentations referencing similar topics