A sample generation method and device

Abstract

The invention discloses a sample generation method and device. The method comprises the steps: inputting data in a pre-collected training set into a neural network model, obtaining a sample generationmodel, and enabling the neural network model to be a neural network model employing an attention mechanism; Inputting pre-collected data in the test set into the sample generation model to generate asample. According to the invention, a neural network model of an attention mechanism is adopted; the neural network model can endow characters in the input data with different weights; Therefore, data information in the input data can be learned more selectively, data with high correlation with the current output data in the input data can be found, a high-precision deep learning model can be finally obtained, training loss is low, and samples generated by the high-precision deep learning model are inevitably higher in legality and diversity.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a sample generation method and device. Background technique [0002] Fuzz testing is a common vulnerability mining technique to discover software vulnerabilities by providing unexpected inputs to the target program and monitoring abnormal results. Among them, samples with high legality and diversity need to be generated in fuzz testing. [0003] The existing sample generation method is to use a neural network with a two-layer Long Short-Term Memory (LSTM) model to train a deep learning model, and then input the test data set provided by the Windows official test team into the trained deep learning model. In the model, samples are obtained. [0004] In the existing sample generation method, the source of test data is provided by the government, which is relatively single, which makes the diversity of the final generated samples poor. Moreover, the LSTM model does not have the abi...

AI Technical Summary

Problems solved by technology

[0005] The present invention provides a sample generation method and device, which can solve the problem of the low accuracy of the deep learning model and the large training loss caused by the neural network training with the two-layer LSTM model in the prior art, which leads to the legality and safety of the final generated samples. The problem of poor diversity

Images