Malicious software family classification avoidance method based on deep reinforcement learning
A reinforcement learning and malware technology, applied in neural learning methods, computer parts, instruments, etc., can solve problems such as large amount of calculation and complex model training process, and achieve the effect of low training cost, easy implementation, and easy implementation.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0048] The present invention will be described in further detail below with reference to the drawings and specific embodiments. Such as figure 1 As shown, a method for evading malware family classification based on deep reinforcement learning of the present invention includes the following steps:
[0049] Step 1: Collect virus samples. The samples are based on Win32 platform from Backdoor, Dos, Email, Exploit, Net-worm, Rootkit, Trojan, Virus, Worm and other malware families in different PE format samples. Use the Python-based lief analysis library to analyze the selected samples, delete the samples with errors in the lief analysis, and complete the data cleaning work. In order to reduce disk IO operations and improve training speed, all samples are cached before training, and all the binary data of the samples are read into the memory. When the status of the file is obtained during the training process, it is directly read from the memory.
[0050] Step 2: Construct an agent (age...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap