My home page
My group
Professional activities

Daphne Koller Publications

Efficient Reinforcement Learning in Factored MDPs (1999)

by M. J. Kearns and D. Koller

Abstract: We present a provably efficient and near-optimal algorithm for reinforcement learning in Markov decision processes (MDPs) whose transition model can be factored as a dynamic Bayesian network (DBN). Our algorithm generalizes the recent E3 algorithm of Kearns and Singh, and assumes that we are given both an algorithm for approximate planning, and the graphical structure (but not the parameters) of the DBN. Unlike the original E3 algorithm, our new algorithm exploits the DBN structure to achieve a running time that scales polynomially in the number of parameters of the DBN, which may be exponentially smaller than the number of global states.

Download Information

M. J. Kearns and D. Koller (1999). "Efficient Reinforcement Learning in Factored MDPs." Proc. Sixteenth International Joint Conference on Artificial Intelligence (IJCAI) (pp. 740-747). pdf ps.gz

Bibtex citation

  author =       "M. J. Kearns and D. Koller",
  booktitle =    "Proc. Sixteenth International Joint Conference on
                 Artificial Intelligence (IJCAI)",
  title =        "Efficient Reinforcement Learning in Factored {MDP}s",
  pages =        "740--747",
  year =         "1999",

full list
Click to go to robotics Click to go to theory Click to go to CS Stanford Click to go to Stanford's Webpage
home | biography | research | papers | my group
courses | professional activities | FAQ | personal