University of Cambridge > Talks.cam > Microsoft Research Cambridge, public talks > Search methods based on Monte-Carlo simulation

Search methods based on Monte-Carlo simulation

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Microsoft Research Cambridge Talks Admins.

Search methods based on Monte-Carlo simulation have revolutionized difficult game-playing domains such as Go and General Game Playing. Monte-Carlo Random Walk planning applies Monte-Carlo ideas to deterministic classical planning. The forward chaining planner Arvand uses such Monte-Carlo random walks to explore the local neighborhood of a search state for action selection. Random walks yield a large unbiased sample of the search neighborhood, and require expensive state evaluations only at the endpoints of each walk. The result is a relatively simple but powerful planner that is competitive with state of the art systems. The talk introduces the main ideas of planning using Monte-Carlo random walks, and briefly discusses recent work to improve plan quality and to improve planning with limited resources.

This talk is part of the Microsoft Research Cambridge, public talks series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2019 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity