Title :
Gibbsboost: a Boosting Algorithm using a Sequential Monte Carlo Approach
Author :
Nakada, Yohei ; Mouri, Yusuke ; Hongo, Yasunori ; Matsumoto, Takashi
Author_Institution :
Grad. Sch. of Sci. & Eng., Waseda Univ., Tokyo
Abstract :
This study proposes a novel boosting algorithm, GibbsBoost. A Gibbs distribution of a weaklearner sequence with a specific loss (energy) function is used in this algorithm as the posterior distribution in Bayesian learning. Weaklearner sequence samples are recursively drawn from the distribution via sequential Monte Carlo. The predictions are derived from a combination of the weaklearner sequence samples. The proposed algorithm is demonstrated by using a numerical example.
Keywords :
Bayes methods; Monte Carlo methods; statistical distributions; Bayesian learning; Gibbs distribution; GibbsBoost; boosting algorithm; energy function; loss function; posterior distribution; sequential Monte Carlo approach; weaklearner sequence; Application software; Bagging; Bayesian methods; Boosting; Computational efficiency; Monte Carlo methods; Power engineering and energy; Robustness; Signal processing algorithms; Sliding mode control;
Conference_Titel :
Machine Learning for Signal Processing, 2006. Proceedings of the 2006 16th IEEE Signal Processing Society Workshop on
Conference_Location :
Arlington, VA
Print_ISBN :
1-4244-0656-0
Electronic_ISBN :
1551-2541
DOI :
10.1109/MLSP.2006.275516
