文章基本信息

标题：A Survey on Various Online Multitask Learning Methodologies-Edification
本地全文：下载
作者：M.Beemamehraj ; Sundararajan.M ; Arulselvi S 等
期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2015
卷号：3
期号：1
DOI：10.15680/ijircce.2015.0301129
出版社：S&S Publications
摘要：Multi-task learning (MTL) is an approach to machine learning that learns a problem together with otherrelated problems at the same time, using a shared representation. This often leads to a better model for the main task,because it allows the learner to use the commonality among the tasks. The major requirement of online applications,isto acheive a highly efficient and scalable problem that can give sudden assumption with low learning cost. Thisrequirement leaves conventional batch learning algorithms out of consideration. Then, novel organization methods, beit group or online, often encounter a dilemma when applied to a group of process, i.e., on one hand, a singleclassification model trained on the entire collection of data from all tasks may fail to capture characteristics ofindividual task; on the other hand, a model trained separately on individual tasks may suffer from insufficient trainingdata. To rectify this problem in this paper, we propose a Edification training for participating in various activitiesthrough online, from this we can geographical model over the entire data of all process.
关键词：on-line learning; Artificial intelligence; multitask learning; classification