Sign in
Combining learned controllers to achieve new goals based on linearly solvable MDPs.
Eiji Uchibe
Kenji Doya
Published in:
ICRA (2014)
Keyphrases
</>
reinforcement learning
markov decision processes
state space
np complete
control system
real time
computational complexity
multi agent systems
model checking
function approximation
controller design
factored mdps