We introduce CUI networks, a compact graphical representation of utility functions over multiple attributes. CUI networks model multiattribute utility functions using the well studied and widely applicable utility independence concept. We show how conditional utility independence leads to an effective functional decomposition that can be exhibited graphically, and how local, compact data at the graph nodes can be used to calculate joint utility and to find the utility maximizing alternative. We discuss the applicability of conditional utility independence compared to other well known independence conditions, and contrast our new representation with previous graphical preference modeling.
