Verifiable privacy protection single-layer perceptron batch training method

Abstract

The invention belongs to the technical field of applying an electronic device to carry out identification and discloses a verifiable privacy protection single-layer perceptron batch training method and a mode identification system. On the basis of the same group of training samples, the training can be carried out for different modes, thereby obtaining a plurality of different training models. Ineach round of iteration, a small batch of samples instead of a sample are selected for iteration. Through utilization of a stochastic gradient descent method, a sample vector is expanded into a matrixsample for acceleration of a convergence rate of the iteration. In a training phase, a user outsources heavy computing tasks to a cloud server. The user needs to carry out encryption operation beforeuploading an input matrix. Through utilization of a stochastic permutation function and a sparse matrix blind technology, the privacy protection of user data is realized. According to the method, a verification mechanism is taken into consideration in a single-layer perceptron training scheme for the first time, the cloud server returns a wrong computing result, and the user can check the wrong computing result with a probability of 100%.

Description

technical field [0001] The invention belongs to the technical field of identification methods or devices using electronic equipment, and in particular relates to a verifiable privacy-protecting single-layer perceptron batch training method. Background technique [0002] At present, the existing technologies commonly used in the industry are as follows: Compared with 2013, the global data volume was 0.9ZT, and by 2020, the data volume will reach 15ZT. With the gradual increase in the amount of data generated by different devices, machine learning solutions have received more and more attention and applications. Machine learning can process massive amounts of data for model training; recently, machine learning has been applied in many research fields. For example: spam classification, disease diagnosis and risk assessment, etc. Machine learning consists of two phases: the training phase and the prediction phase. Given a set of training examples and corresponding output val...

AI Technical Summary

Problems solved by technology

[0004] (1) The existing training model can only complete the training of one model at a time, and cannot realize the training task of multiple models;

[0005] (2) In model training, each time a sample is selected for iterative update, the iterative convergence speed is slow;

[0006] (3) The existing encryption technology can realize the privacy protection of user data, but there is currently a lack of efficient and feasible privacy protection machine learning solutions;

[0007] (4) The existing cloud server-based model training has no verification function. Since the cloud server is usually semi-trusted, the cloud server may return invalid calculation results, which may cause the training task to fail

Images