Privacy Conserving Gradient Descent Method Applied for Neural Network with Distributed Datasets

Sachin P. Yadav, Amit B. Chougule

Abstract: The learning problems have to be concerned about distributed input data, because of gradual expansion of distributed computing environment. It is important to address the privacy concern of each data holder by extending the privacy preservation concept to original learning algorithms, to enhance co-operations in learning. In this project, focus is on protecting the privacy in significant learning model i. e. Multilayer Back Propagation Neural Network using Gradient Descent Methods. For protecting the privacy of the data items (concentration is towards Vertically Partitioned Data and Horizontally Partitioned Data), semi honest model and underlying security of El Gamal Scheme is referred [7].

Keywords: Cryptography Techniques, Distributed Datasets, Gradient Descent Methods, Back Propagation Neural Network

How to Cite?: Sachin P. Yadav, Amit B. Chougule, "Privacy Conserving Gradient Descent Method Applied for Neural Network with Distributed Datasets", Volume 4 Issue 6, June 2015, International Journal of Science and Research (IJSR), Pages: 1592-1597, https://www.ijsr.net/getabstract.php?paperid=SUB155632, DOI: https://dx.doi.org/10.21275/SUB155632