期刊名称：Lecture Notes in Engineering and Computer Science
印刷版ISSN：2078-0958
电子版ISSN：2078-0966
出版年度：2019
卷号：2239
页码：1-6
出版社：Newswood and International Association of Engineers
摘要：Cloud (computing) system has been widely used as
one of ICTs. However, as the number of terminals connected to
it increases, the limit of the capability is also becoming apparent.
The limit of its capability leads to the delay of significant
processing time. In order to improve this, the edge (or fog)
computing system has been proposed. In the conventional cloud
system, a terminal sends all data to the cloud and the cloud
returns the computation result to the terminal (or thing) directly
connected to it. On the other hand, in the edge (computing)
system, a plural of servers called edges are assigned between the
cloud and the terminal (or thing). In the system, there are two
groups of servers for cloud and edge. Heavy and normal tasks
are processed in cloud and edge servers, respectively. Then, let
us consider about machine learning in cloud or edge system. The
purpose of learning is to find out the relationship (information)
lurking in from the collected data. That is, a system with
several parameters is assumed and estimated by repeatedly
updating the parameters with learning data. Further, there is
the problem of the security for learning data. How can we build
cloud system to avoid such risk? Secure multiparty computation
(SMC) is known as one method realizing secure computation.
Many studies on learning methods based on SMC have also
been proposed in the cloud system. Then, what kind of learning
method is suitable for edge system based on SMC? In this
paper, we will propose Neural Gas (NG) algorithms to realize
fast and secure processing on edge computing for clustering
and classification problems, and show the effectiveness of the
proposed methods in numerical simulations.
