A new noise robust ensemble method called "Averaged Boosting (A-Boosting)" is proposed. Using the hypothetical ensemble algorithm in Hilbert space, we explain that A-Boosting can be understood as a method of constructing a sequence of hypotheses and coefficients such that the average of the product of the base hypotheses and coefficients converges to the desirable function. Empirical studies showed that A-Boosting outperforms Bagging for low noise cases and is more robust than AdaBoost to label noise.
CITATION STYLE
Kim, Y. (2003). Averaged Boosting: A noise-robust ensemble method. In Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science) (Vol. 2637, pp. 388–393). Springer Verlag. https://doi.org/10.1007/3-540-36175-8_38
Mendeley helps you to discover research relevant for your work.