Approximate dynamic programming with affine ADDs

  title={Approximate dynamic programming with affine ADDs},
  author={Scott Sanner and William T. B. Uther and Karina Valdivia Delgado},
The Affine ADD (AADD) is an extension of the Algebraic Decision Diagram (ADD) that compactly represents context-specific , additive and multiplicative structure in functions from a discrete domain to a real-valued range. In this paper, we introduce a novel algorithm for efficiently finding AADD approximations that we use to develop the MADCAP algorithm for AADD-based structured approximate dynamic programming (ADP) with factored MDPs. MADCAP requires less time and space to achieve comparable or… CONTINUE READING


Publications referenced by this paper.
Showing 1-10 of 12 references

Puterman.Markov Decision Processes: Discrete Stochastic Dynamic Programming

  • M L.
  • 1994
Highly Influential
4 Excerpts

Effici ent solution methods for factored MDPs

  • C. Guestrin, D. Koller, R. Parr, S. Venktaraman
  • Journal of Artificial Intelligence Research (JAIR…
  • 2003
2 Excerpts

Similar Papers

Loading similar papers…