@INPROCEEDINGS{RaeOnoMue98b,
    author = {R{\"a}tsch, G. and Onoda, T. and M{\"u}ller, K.--R.},
    editor = "Kearns, M.S. and Solla, S.A. and Cohn, D.A.",
    title = "Regularizing {A}da{B}oost",
    booktitle = "Proc. NIPS 11",
    year = "1999",
    pages = "564--570",
    publisher = "MIT Press",
    abstract = "Also for noisy data boosting will try to enforce a hard margin and thereby give too much weight to outliers, which then leads to the dilemma of non-smooth decision surfaces and overfitting. Therefore we propose three algorithms to allow for soft margin classification by introducing regularization with slack variables into the boosting concept: (1) AdaBoost$\_{reg}$ and regularized versions of (2) linear and (3) quadratic programming AdaBoost. Experiments show the usefulness of the proposed algorithms in comparison to another soft margin classifier: the support vector machine.",
    pdf = "http://ida.first.fhg.de/\\textasciitilde {}raetsch/ps/RaeOnoMue98d.pdf",
    postscript = "http://ida.first.fhg.de/\\textasciitilde {}raetsch/ps/RaeOnoMue98d.ps.gz"
}