Partial-monitoring games are a mathematical framework for sequential decision
problems with imperfect feedback: The learner repeatedly chooses an action, the
nature responds with an outcome, and then the learner suffers a loss and
receives a feedback signal, both of which are fixed functions of the action and
the outcome. The goal of the learner is to minimize his total cumulative loss.
We make progress towards classification of these games based on their minimax
expected regret.