Handling Big Data in set-membership identification through a sparse optimization approach

详细信息    查看全文

• 作者：V. Cerone^* ; ^{vito.cerone@polito.it" class="auth_mail" title="E-mail the corresponding author} ; D. Regruto^* ; ^{diego.regruto@polito.it" class="auth_mail" title="E-mail the corresponding author}
• 关键词：Big Data ; Set-membership identification
• 刊名：IFAC-PapersOnLine
• 出版年：2015
• 出版时间：2015
• 年：2015
• 卷：48
• 期：28
• 页码：1272-1278
• 全文大小：446 K

文摘

The tutorial paper discusses how to handle a large amount of data in a set- membership (SM) identification problem. First, the meaning of the term "Big data" will be clarified and explained in the considered context and, in particular, it will be shown how even a few dozen of input-output data could be "Big enough" to prevent practical estimation of the model due to the complexity of the involved optimization problems. Then, possible ad hoc solutions to properly handle the availability of large amount of data will be presented with particular attention to the peculiar sparsity structure of the considered estimation/identification problems. Numerical examples will be presented.