New Privacy Preserving Back Propagation Learning for Secure Multiparty Computation

Many studies have been done with the security of cloud computing. Data encryption is one of typical approaches. However, complex computing requirement for encrypted data needs a great deal of time and effort for the system in this case. Therefore, another studies on secure sharing and computing methods are made to avoid secure risks being abused or leaked and to reduce computing cost. The secure multiparty computation (SMC) is one of these methods. So far, some studies have been done with SMC. Specifically, SMC with secure shared data in addition and multiplication forms is proposed and applied to arithmetic operation and simple statistical computation. However, complex calculation processing such as BP(Back Propagation) learning has never proposed yet. In this paper, we propose BP learning method for SMC on cloud computing system and prove the validity of it. Further, the performance of the proposed method is shown in numerical simulations.