Over the past decade, the rise of machine learning (ML) and deep learning (DL) evolved in various life areas, especially medical, cyber security, finance, and education. This has dramatically increased the attack surface for the vibrantly used neural network venerable to so-called adversarial attacks. On the other hand, new threats are also being discovered daily, making it harder for current solutions to cope with a large amount of data to analyse. Numerous machine learning algorithms have found their ways in the mentioned fields to identify new and unknown attacks.

While these applications of machine learning algorithms have been proven beneficial in various fields, they have also highlighted many shortcomings, such as the lack of datasets, the inability to learn from small datasets, the cost of the architecture, and imbalanced datasets name a few. On the other hand, new and emerging algorithms, such as Deep Learning, One-shot Learning, Continuous Learning and Generative Adversarial Networks, have been successfully applied to solve various tasks in these fields. Therefore, it is crucial to apply these new methods to life-critical missions and measure these less-traditional algorithms' success when used in these fields.