1 2 When studying the training error and the prediction error for boosting, it is often assumed that the hypotheses returned by the base learner are weakly accurate, or are able to beat a random guesser by a certain amount of diierence. It is has been an open question how much this diierence can be, whether it will eventually disappear in the boosting process or be bounded by a nite amount see,...

Constraint satisfaction problems are an important class of problems in complexity theory. They generalize many combinatorial problems as well as satisfiability problems and provide canonical complete problems for many complexity classes. The computational complexity of all Boolean constraint satisfaction problems was classified by Schaefer [Sch78] and reveals a dichotomic behavior that is conje...