[brockf]

It's great to see more hacker-friendly introductions to reinforcement learning. Like most facets of machine learning, there are so many interesting applications of reinforcement learning (e.g., we're using RL to optimize email marketing campaigns at Optimail), and we'll only find more as more non-academic hackers discover it.

[make3]

... still, I think it does people disservice to not make it clear that ml is math and that it can only really be done well if you understand what is happening, ie, if you take the time to understand the maths, which is not that hard btw.

Calling something cool/hacking because you don't want to take the time to understand the maths is something Trump would do if he was a programmer

[spenuke]

Can you expand on what math you're talking about that is both necessary and not that hard? "Not that hard", to me, indicates that a reasonably intelligent person could teach themselves without university instructors (present or past).