Running summarizing tools on a new article

Abstract: Abstract Adversarial examples are the latest cyber threats endangering networking systems. An adversarial example is an input to a machine learning model whose goal is to cause a misclassification to occur after applying a small perturbation to the original input. This type of adversarial example has been impacting multiple intelligent domains, including recognition of image, audio, video samples, and critical security applications such as intrusion detection and malware detection. Machine learning models are extremely vulnerable to attacks using adversarial examples. Machine learning models that classify Windows Portable Executable (PE)files are challenging to attack using existing methods that generate adversarial examples. This chapter introduces the concept of adversarial examples and its challenges. They are followed by three research studies from both offensive and defensive perspectives. The first study is to discover attack methods that are much more sophisticated and difficult to detect than the current methods that simply append large amounts of specially crafted byte sequences to the end of the PE file. The second study proposes a metric for quantizing the vulnerability or susceptibility of a given machine learning model to adversarial manipulation. The third study presents directions on protecting smart speakers from adversarial voice commands.