Backdoor confrontation sample generation method of PE malicious software detection model

Abstract

The invention relates to a backdoor confrontation sample generation method of a PE malicious software detection model based on R-DBSCAN, and belongs to the field of computer malicious software detection. The method mainly aims at solving the problem that a malicious software detection model is high in attack difficulty under the black box condition. The method comprises the following steps: firstly, acquiring a PE sample from a public data set, training a proxy training model, and reducing the dimension of the data set by adopting an SHAP value; clustering the samples by adopting an R-DBSCAN method, and taking a center node of each cluster as a sampling point to construct a new data set; training a neural network model; respectively inputting malicious and benign sample files, and recording neurons which greatly influence a classification result according to the weight change condition of the neurons in the neural network; embedding a character string with any length into the empty PE file, taking a character string which greatly influences the character string according to the weight change condition of the neuron, and recording the neuron; embedding a trigger into an original malicious PE file, and modifying a label to achieve adversarial training of a neural network.

Description

Technical field [0001] The present invention relates to a rear door counterpart generation method of a PE malware detection model, belonging to the field of computer malware detection. Background technique [0002] Testing techniques for malware have always had a rapid development, but the number of new numbers of malware is still very considerable. With the development of deep learning, it becomes a variety of PE malware detection methods based on deep learning. Static analysis of PE files, extracts document 16 Enciprome data characteristics, use depth learning methods learning characteristics, this type of method is currently able to achieve detection of malware with higher accuracy. However, such malware detection models ignore the safety of the detection system itself and the reliability of the data itself during the development process. If the malware detection model encounters an attacker inserted into the back door during the training process, the detection model will guar...

AI Technical Summary

Problems solved by technology

[0003] At present, the generation method of adversarial samples for PE malware has the characteristics of poor generalization ability and high computational overhead. The construction of adversarial samples is generally designed for a specific detection model, which limits the generalization ability of adversarial samples on other detection models.

The construction of black-box adversarial samples requires a large number of query operations on the detection model to clarify the deceptive effect of the adversarial samples on the model. In practice, a large number of queries are time-consuming and laborious, which reduces the practicability of the adversarial sample generation method.

Images