Uncertainty in Artificial Intelligence
First Name   Last Name   Password   Forgot Password   Log in!
    Proceedings   Proceeding details   Article details         Authors         Search    
UCP-Networks: A Directed Graphical Representation of Conditional Utilities
Craig Boutilier, Fahiem Bacchus, Ronen Brafman
Abstract:
We propose a new directed graphical representation of utility functions, called UCP-networks, that combines aspects of two existing graphical models: generalized additive models and CP-networks. The network decomposes a utility function into a number of additive factors, with the directionality of the arcs reflecting conditional dependence of preference statements---in the underlying (qualitative) preference ordering---under a {em ceteris paribus} (all else being equal) interpretation. This representation is arguably natural in many settings. Furthermore, the strong CP-semantics ensures that computation of optimization and dominance queries is very efficient. We also demonstrate the value of this representation in decision making. Finally, we describe an interactive elicitation procedure that takes advantage of the linear nature of the constraints on ``tradeoff weights'' imposed by a UCP-network. This procedure allows the network to be refined until the regret of the decision with minimax regret (with respect to the incompletely specified utility function) falls below a specified threshold (e.g., the cost of further questioning.
Keywords:
Pages: 56-64
PS Link:
PDF Link: /papers/01/p56-boutilier.pdf
BibTex:
@INPROCEEDINGS{Boutilier01,
AUTHOR = "Craig Boutilier and Fahiem Bacchus and Ronen Brafman",
TITLE = "UCP-Networks: A Directed Graphical Representation of Conditional Utilities",
BOOKTITLE = "Proceedings of the Seventeenth Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-01)",
PUBLISHER = "Morgan Kaufmann",
ADDRESS = "San Francisco, CA",
YEAR = "2001",
PAGES = "56--64"
}


hosted by DSL   •   site info   •   help