The Convergence Rate of AdaBoost

Abstract. We pose the problem of determining the rate of convergence at which AdaBoost minimizes exponential loss. Boosting is the problem of combining many "weak," high-error hypotheses to generate a single "strong"

• Type: Journal
• Year: 2010
• Where: COLT
• Authors: Robert E. Schapire
• Keywords: AdaBoost, COLT 2010, Convergence, Exponential Loss, Machine Learning