Hacker News new | ask | show | jobs
by ipaddr 500 days ago
Now you are asking for a perfect modeling of the system. Reinforcement learning works by discovering boundaries.