Hackers have developed a toolkit for developing and comparing reinforcement learning algorithms. This makes it possible to write agents that learn to manipulate PE files (e.g., malware) to achieve some objective. Around 60% of samples produced can bypass the system security. In a span of 15 hours of training, the software ran over 100,000 samples past an unnamed security engine. The code for OpenAI Gym was published in Github by Hyrum Anderson and their team. The software was developed by Endgame security shop Endgame.”]
Source: https://gbhackers.com/machine-learning-system-to-create-invisible-malwares-gym-malware/