|
|
|
|
|
|
by gdb
3180 days ago
|
|
|
(I work at OpenAI.) Correct, we've been playing a number of pros using the same bot played at TI. We do have a stronger version which is just two days more of training (gets a 70% win rate vs the one at TI), but haven't seen a need to test it out. We'll likely do a blog post in upcoming weeks with more stats and commentary from the pros; would be curious what people would like to know! Incidentally, the various exploits that people used are all similar to how we actually develop the bot. We try to find areas of the strategy space it hasn't explored, and then make a small tweak to encourage it to explore that area. Lots of progress comes from removing hardcoded restrictions, which are nice to get started. So the fact there exist exploits wasn't surprising to us — what would be surprising would be exploits we couldn't fix. 1v1 has always been a proof-of-concept for us. The fun part is really 5v5, which is what we're working on now (and hiring for! ping me if you're interested: gdb@openai.com). |
|
|
I understand this is a perfectly reasonable early test, but there are so many complaints about "it was just a restricted subset of the game and 1v1".
This is like complaining that Google doesn't release first-pass code (with minimal unit tests and no stress testing) to their production sites across the world. Everything that loops starts with the first iteration.
Also, keep up the good work, OpenAI! And please remember Asimov's 3 rules.