稀疏学习的对偶增强拉格朗日算法的超线性收敛性Super-Linear Convergence of Dual Augmented Lagrangian Algorithm for Sparse Learning |
|
课程网址: | http://videolectures.net/nipsworkshops09_tomioka_slc/ |
主讲教师: | Ryota Tomioka |
开课单位: | 芝加哥丰田技术学院 |
开课时间: | 2010-01-19 |
课程语种: | 英语 |
中文简介: | 分析了一种新提出的稀疏学习算法——双增广拉格朗日算法(DAL)的收敛性。我们从理论上分析了DAL在非渐近全局意义上超线性收敛的条件。在大尺度_1-正则化逻辑回归问题上,实验验证了本文的分析方法,并将DAL算法与现有算法进行了比较。 |
课程简介: | We analyze the convergence behaviour of a recently proposed algorithm for sparse learning called Dual Augmented Lagrangian (DAL). We theoretically analyze under some conditions that DAL converges super-linearly in a non-asymptotic and global sense. We experimentally confirm our analysis in a large scale ℓ1-regularized logistic regression problem and compare the efficiency of DAL algorithm to existing algorithms. |
关 键 词: | 计算机科学; 优化方法; 稀疏学习 |
课程来源: | 视频讲座网 |
最后编审: | 2020-06-03:毛岱琦(课程编辑志愿者) |
阅读次数: | 45 |