Search

With the prosperity of artificial intelligence, research on machine learning becomes a hot issue globally. Generative Adversarial Networks expose the huge security risks of machine learning. After the creation, GAN has achieved good performance in image generation, automatic image coloring, and data enhancement. With the improvement of the ability to generate samples against the deep learning network, generating malicious samples against the target learning model to achieve the deceptive discriminator becomes an effective and harmful attacking method. At present, some efficient attack methods have been proposed for different types of learning networks and different types of sample data. This paper mainly discusses the vulnerability of deep learning networks and several attack methods based on adversarial samples.

In these years, with the development of search engines, it is more convenient for people to get information via the Internet. However, despite the benefit, search engines bring new challenges to network security. More and more hackers search for vulnerable websites and computers with the help of search engines and make large-scale attacks. This is quite a challenge to traditional cybersecurity concepts. Honeypot is a technology that is widely used in capturing and analyzing hackers' behaviors. By building virtual objectives with vulnerabilities, honeypots attract hackers and analyze the interaction process with hackers. But traditional honeypot can't attract hackers well and just wait for attackers. We design and implement a new Web honeypot detection system based on search engines, which can not only attract hackers but also can renew the content automotive to synchronize with the new attack technologies. This new kind of honeypot makes up for the shortcomings of traditional honeypots and has great advantages in finding zero-day attacks.