312 search results - page 5 / 63 » Learning Partially Observable Deterministic Action Models |
ICML
2009
IEEE
14 years 8 months ago
2009
IEEE
Large margin learning of Continuous Density HMMs with a partially labeled dataset has been extensively studied in the speech and handwriting recognition fields. Yet due to the non...
IJCAI
2007
13 years 9 months ago
2007
The association of perception and action is key to learning by observation in general, and to programlevel task imitation in particular. The question is how to structure this info...
ICML
2008
IEEE
14 years 8 months ago
2008
IEEE
Partially Observable Markov Decision Processes (POMDPs) have succeeded in planning domains that require balancing actions that increase an agent's knowledge and actions that ...
DSN
2006
IEEE
14 years 1 months ago
2006
IEEE
This paper provides a technique, based on partially observable Markov decision processes (POMDPs), for building automatic recovery controllers to guide distributed system recovery...
ECML
2005
Springer
14 years 1 months ago
2005
Springer
This paper examines the problem of finding an optimal policy for a Partially Observable Markov Decision Process (POMDP) when the model is not known or is only poorly specified. W...