透過您的圖書館登入
IP:18.224.64.24
  • 期刊

Research on Privacy and Security of Federated Learning in Intelligent Plant Factory Systems

摘要


Intelligent plant factories must continuously collect sensing data from other factories to update parameters in real time and obtain the best plant growth environment. However, if the collected sensing data is forged or collected, it will affect the performance of machine learning. In this paper, after using the federated learning technology to train the local data in each region, upload the gradient loss and generate blocks in the blockchain. Finally, verify and aggregate it into a complete model trained with all the data instead of directly uploading the local data to ensure privacy and data security.

延伸閱讀