Agents learned to navigate a environment (random initial parameters, Brownian noise in all environment states). However, they still haven't figured how to carry out a successful flash loan attack yet! https://tinyurl.com/mangoooo0
Project Mango
Training an RL agent to autonomously discover DeFi exploits, inspired by the $114M Mango Markets hack.
Log in or sign up for Devpost to join the conversation.