International Journal of Innovative Research in Science, Engineering and Technology (IJIRSET)
During the collaborative learning of neural network on cloud computing, multiple parties collaborate with their own private data sets. As the collaborative neural network learning process starts an individual datasets of all parties are collected and during this learning process no party needs to disclose the users' non-public information to others. The existing methods that support this kind of approach for collaborative learning, but these are limited only for two parties as well as data partition. There is no solution for two or more parties, for collaboratively conduct the learning, with an arbitrarily partitioned data set. Here is a solution presented to overcome the above stated limitations of existing collaborative learning methods used in cloud computing.