CAJ | 학술논문

MIRA(Margin Infused Relaxed Algorithm)是一种超保守算法,在分类、排序、回归等应用领域都取得不错成绩.文中在传统MIRA算法基础上进行改进,提出k-best MIRA(K-MIRA)与动态k-best MIRA(DK-MIRA)算法.这两种算法能够根据学习进程自动调整优化约束条件,从而提高算法的收敛速度与性能.将K-MIRA与DK-MIRA用于定义类问题回答中的句子排序任务,取得较为满意的实验结果.
MIRA(Margin Infused Relaxed Algorithm)시일충초보수산법,재분류、배서、회귀등응용영역도취득불착성적.문중재전통MIRA산법기출상진행개진,제출k-best MIRA(K-MIRA)여동태k-best MIRA(DK-MIRA)산법.저량충산법능구근거학습진정자동조정우화약속조건,종이제고산법적수렴속도여성능.장K-MIRA여DK-MIRA용우정의류문제회답중적구자배서임무,취득교위만의적실험결과.
Margin infused relaxed algorithm (MIRA) is an improved ultraconservative algorithm, which is successfully used in classification, ranking and regression. The k-best MIRA (K-MIRA) and dynamic k-best MIRA (DK-MIRA) axe proposed. The improved MIRA reduces the optimization constraints progressively as training moves forward. The experiment is carried out on the task of sentence ranking in definitional question answering with K-MIRA and DK-MIRA. The experimental results show that the proposed algorithms greatly improve the performance.