Sciweavers

Share
NIPS
2008
9 years 5 months ago
Particle Filter-based Policy Gradient in POMDPs
Our setting is a Partially Observable Markov Decision Process with continuous state, observation and action spaces. Decisions are based on a Particle Filter for estimating the bel...
Pierre-Arnaud Coquelin, Romain Deguest, Rém...
books