24 search results - page 2 / 5 » Learning Policy Improvements with Path Integrals |
SOSP
2007
ACM
14 years 2 months ago
2007
ACM
We introduce a new reliability infrastructure for file systems called I/O shepherding. I/O shepherding allows a file system developer to craft nuanced reliability policies to de...
IJCAI
2003
13 years 6 months ago
2003
We investigate the problem of non-covariant behavior of policy gradient reinforcement learning algorithms. The policy gradient approach is amenable to analysis by information geom...
LCTRTS
2007
Springer
13 years 11 months ago
2007
Springer
Embedded systems serve an emerging and diverse set of applications. As a result, more computational and storage capabilities are added to accommodate ever more demanding applicati...
IJCAI
2007
13 years 6 months ago
2007
Many current state-of-the-art planners rely on forward heuristic search. The success of such search typically depends on heuristic distance-to-the-goal estimates derived from the ...
ECAI
2006
Springer
13 years 9 months ago
2006
Springer
Abstract. This paper proposes a novel approach to discover options in the form of conditionally terminating sequences, and shows how they can be integrated into reinforcement learn...