IPBoost -- Non-Convex Boosting via Integer Programming

Boosting is an important (and by now standard) technique in classification to combine several'low accuracy' learners, so-called base learners, into a'high accuracy' learner, a so-called boosted learner. Pioneered by the AdaBoost approach of [19], in recent decades there has been extensive work on boosting procedures and analyses of their limitations. In a nutshell, boosting procedures are (typically) iterative schemes that roughly work as follows: for t 1,..., T do the following: