Design of Distributed Terminal Data Joint Learning Model Based on Blockchain

With the rapid development of new-generation information technologies people enjoy the convenience of technology and are also troubled by privacy data leakage. Especially in artificial intelligence technology, its development is inseparable from the support of data. The quality of data directly determines the effect of the model. However, in recent years, with the increasing awareness of privacy protection and the promulgation of relevant laws and regulations, the collection and use of data has become a major problem, which greatly limits the development of artificial intelligence technology. The emergence of federated learning needs to solve this problem. It ensures that the data can not be exported to the user's local area and can also achieve the purpose of data utilization. This paper proposes a federated learning model based on the blockchain decentralized architecture, which aims to solve the contradiction between data privacy protection and data utilization, and achieve the purpose of data “available invisible”; finally, three different experiments were tested. The training method is summarized.