一种利用Screening加速技巧的Lasso算法-计算机工程与应用.PDFVIP

Computer Engineering and Applications 计算机工程与应用 2018 ，54（4 ） 135 一种利用Screening 加速技巧的Lasso 算法 用 邱俊洋，潘志松，易 磊，陶 蔚，张梁梁 应 QIU Junyang, PAN Zhisong, YI Lei, TAO Wei, ZHANG Liangliang 与 中国人民解放军理工大学 指挥信息系统学院，南京 210007 程 College of Command Information System, PLA University of Science and Technology, Nanjing 210007, China g 工 r QIU Junyang, PAN Zhisong, YI Lei, et al. Lasso algorithm using Screening accelerated technique. Computer Engi- o . neering and Applications, 2018, 54 （4 ）：135-140. j 机 a e Abstract ：Recently, Lasso is one of the most popular algorithms for sparse feature selection in the application. Traditional 算 c Lasso algorithm reduces the computational overhead to a certain extent through feature selection. However, sometimes it may . become impossible to load the data matrix into the main memory of the computer when the feature dimension or the number 计 w of samples is extremely large. Thus it can be seen that there are still many difficulties and challenges in solving Lasso algo- w rithm. To address these challenges, the Screening accelerating technique has been proposed and becomes the research fo- w cus in recent years. The Screening technique is able to quickly identify and remove the invalid features before the optimi- zation process. Then it only needs to deal with the reduced data, which can greatly accelerate the solving efficiency. The most important is that the Screening technique can guarantee that the discarded features have zero coeff