Bounding Performance Loss in Approximate MDP Homomorphisms.

Jonathan Taylor Doina Precup Prakash Panangaden

Published in: NIPS (2008)

Keyphrases

markov decision processes
upper bound
markov decision process
factored mdps
state space
optimal policy
finite state
utility function
reinforcement learning
data sets
least squares
linear program
graph theory
information loss
machine learning
planning under uncertainty
finite state automata