Federal learning method and device, equipment and storage medium

Abstract

The invention provides a federal learning method and device, equipment and a storage medium, and the method comprises the steps: building an initial model related to a block chain node, and enabling the initial model to serve as a local model of the block chain node; performing model training on the local model by using a meta-learning method in combination with local data of the block chain node to obtain an updated local model; and performing model training on the updated local model in combination with local data of the block chain node until the local model is converged to a preset range, thereby determining a result model of the block chain node. According to the method, through the meta-learning algorithm, the optimal solution can be quickly converged only through a small amount of training, the initial model can ensure that the initial model can be quickly converged on later similar tasks, the corresponding result model is obtained, computing resources are saved, and benign operation of the system is facilitated.

Description

technical field [0001] The present application relates to the field of communication technology, and in particular to a federated learning method, device, computer equipment and storage medium. Background technique [0002] Federated Learning (Federated Learning) was first proposed by Google. It is to carry out high-efficiency learning among multiple participants or multiple computing nodes under the premise of ensuring information security during data exchange, protecting terminal data and personal data privacy, and ensuring compliance with laws and regulations. It is an emerging artificial intelligence basic technology for machine learning. Unlike distributed machine learning, federated learning requires a decentralized distribution system to ensure the privacy and security of users. It is effective under the premise of ensuring data security, exchange, and training efficiency. machine learning. The federated learning model is trained and uploaded locally by each device, ...

AI Technical Summary

Problems solved by technology

[0004] The encryption and decryption process of the blockchain, and the proof mechanism, etc., the complexity of these processes will greatly reduce the efficiency of model training

For more complex models, the encryption and transmission of model parameters will take a long time

The excessive size of the model in the iterative process stored in the blockchain will also lead to a large storage cost

Images