Distributed deep learning training method based on mobile device for protecting data privacy

Abstract

The invention discloses a distributed deep learning training method based on a mobile device for protecting data privacy. The method comprises the following steps of 1, cutting a neural network; 2, distributing the mobile device; 3, training in a mobile device group; 4, training among the mobile device groups; and 5, updating a global model of a global server. According to the distributed deep learning training method, the purposes of protecting the data privacy of the user and enabling the training model to achieve high performance are achieved, the user enjoys the absolute control right for the training data generated by the user, and the problem that a service provider can maliciously use, violate and leak the user data privacy under the traditional centralized deep learning, is solved.

Description

Technical field [0001] The present invention relates to the field of deep learning, in particular to a distributed deep learning training method based on mobile equipment that protects data privacy. Background technique [0002] The application of deep learning in many fields, such as speech recognition, object recognition, face detection, and biomedicine, has far exceeded the performance of traditional machine learning algorithms. Machine learning is good at capturing nonlinear features from complex data structures, and at the same time has strong robustness to uncorrelated noise. The ability of deep learning to achieve such high performance is extremely dependent on massive and diverse training data and complex neural network structures. The collected training data is required to be consistent with the data distribution in actual use. The common practice of service providers at this stage is to implicitly collect data generated by users in their daily use as training data, an...

AI Technical Summary

Problems solved by technology

The existing technology proposes a similar distributed deep learning training model, but fails to fully consider the computing and storage performance of mobile devices

Traditional deep learning training usually needs to be trained for several days on several servers with extremely strong performance. If the computing performance of mobile devices is not optimized, distributed deep learning models cannot be trained on mobile devices.

At the same time, considering that the neural network is becoming more and more complex, a neural network may take up hundreds of megabytes or even gigabytes of storage space, and directly storing the neural network model on the mobile device will greatly consume the precious storage of the mobile device resource

Existing technologies also protect the privacy of training data for deep learning. Common methods such as homomorphic encryption or differential privacy, but homomorphic encryption methods will increase extremely high computational overhead, and because they only support a limited number of additions and multiplications, leading to a sharp drop in the performance of the trained model

The method based on differential privacy will introduce more noise, and the model performance will be poor, so it is not suitable for commercial products.

Images