Advances in machine learning in recent years have created a huge number of applications such as applied data analysis, voice assistants or self-driving cars. The success of machine learning is ensured by the fact that the same methods, in different wrappers, work well in completely different tasks. This allows you to replace the classical methods, gaining in quality and speed of work.
. . Google , Android, . Amazon harvest.AI Macie — , , , .
, , , .
, :
. .
DDoS-. , , .
, . , —
Darktrace. " ", . Ebay, Samsung Micron.
, Darktrace "" , . DarkTrace-antigena — , .
, , , Darktrace , , “ " .” , WannaCry 2017.
, . ImmuniWeb, Intruder .
, . . , .
IBM , , , - , . , 100 .
Ubisoft , . CLEVER - , , Ubisoft , . 79% , . , .
70% Ubisoft – .
, , , . , , . .
, 2016 (CSAIL MIT) AI2 — , " ". , , .
, , , . , , SVD RNN, .
. . , CSAIL, , 85 .
This is just a small list of the improvements machine learning can offer for cybersecurity. It can help in various applications, namely, detecting attacks, finding vulnerabilities in the code, and helping to analyze large amounts of data. I am sure that the potential of machine learning in the field of computer security has not yet been fully explored, and more interesting new applications await us.