Given a set of observed economic choices, can one infer
preferences and/or utility functions for the players that are
consistent with the data? Questions of this type are called {\em
rationalization} or {\em revealed preference} problems in the
economic literature, and are the subject of a rich body of work.
