Live data from Hacker News

OpenAI at the Dota 2 World Championships

openai.com

61–70 of 219 posts

Re: OpenAI at the Dota 2 World Championships

#61

These types of MOBA games are a good precursor of military unit management. To be honest, I'm not totally sure whether to be happy or sad about this sort of thing. The strategic control of drone units on a combat field, without the loss of personel (on the countries with this technology, anyways) should make me happy, and does, to a point. BUT the potential for abuse (and, no, I haven't even begun to extrapolate towa…

Yes, this is perfect for a future of swarms of drones fighting it out against each other.

Re: OpenAI at the Dota 2 World Championships

#62

These types of MOBA games are a good precursor of military unit management. To be honest, I'm not totally sure whether to be happy or sad about this sort of thing. The strategic control of drone units on a combat field, without the loss of personel (on the countries with this technology, anyways) should make me happy, and does, to a point. BUT the potential for abuse (and, no, I haven't even begun to extrapolate towa…

DotA is not a MOBA and it's not a very good precursor for military unit management. I'd say Starcraft is a better comparison, and they are working on applying DeepMind to that right now [0].

[0] https://techcrunch.com/2017/08/09/blizzard-and-deepmind-turn...

Re: OpenAI at the Dota 2 World Championships

#63
post #54

The developer interviewed claimed that there was no domain-specific knowledge and implied (though didn't explicitly state) there wasn't any training against non-OpenAI bots or players. (I'd love to know the reward function they used for whatever Q-learning variation they ran with.) If this is accurate, one of the things I'm most impressed with is that the bot figured out creep-blocking. (I can't find a good GIF, but…

I expect your conjecture is correct. That's the whole point of deep learning -- there are many layers that automate what would otherwise be human feature extraction.

Re: OpenAI at the Dota 2 World Championships

#65
post #62

These types of MOBA games are a good precursor of military unit management. To be honest, I'm not totally sure whether to be happy or sad about this sort of thing. The strategic control of drone units on a combat field, without the loss of personel (on the countries with this technology, anyways) should make me happy, and does, to a point. BUT the potential for abuse (and, no, I haven't even begun to extrapolate towa…

DotA is not a MOBA and it's not a very good precursor for military unit management. I'd say Starcraft is a better comparison, and they are working on applying DeepMind to that right now [0]. [0] https://techcrunch.com/2017/08/09/blizzard-and-deepmind-turn...

DotA and Dota 2 are the games that helped define the term MOBA. I'm not sure why you're saying it isn't one, especially since it's in the first sentence of both Wikipedia articles.

Re: OpenAI at the Dota 2 World Championships

#66
post #62

These types of MOBA games are a good precursor of military unit management. To be honest, I'm not totally sure whether to be happy or sad about this sort of thing. The strategic control of drone units on a combat field, without the loss of personel (on the countries with this technology, anyways) should make me happy, and does, to a point. BUT the potential for abuse (and, no, I haven't even begun to extrapolate towa…

DotA is not a MOBA and it's not a very good precursor for military unit management. I'd say Starcraft is a better comparison, and they are working on applying DeepMind to that right now [0]. [0] https://techcrunch.com/2017/08/09/blizzard-and-deepmind-turn...

DotA is generally considered to have created the MOBA genre. That being said, you are right that MOBA's aren't as good a comparison for military unit management as games like Starcraft or Warcraft.

Re: OpenAI at the Dota 2 World Championships

#67
post #62

These types of MOBA games are a good precursor of military unit management. To be honest, I'm not totally sure whether to be happy or sad about this sort of thing. The strategic control of drone units on a combat field, without the loss of personel (on the countries with this technology, anyways) should make me happy, and does, to a point. BUT the potential for abuse (and, no, I haven't even begun to extrapolate towa…

DotA is not a MOBA and it's not a very good precursor for military unit management. I'd say Starcraft is a better comparison, and they are working on applying DeepMind to that right now [0]. [0] https://techcrunch.com/2017/08/09/blizzard-and-deepmind-turn...

Dota definitely is a MOBA. Dota is the MOBA from which all the others are devised [0].

[0]: https://en.wikipedia.org/wiki/Multiplayer_online_battle_aren...

Re: OpenAI at the Dota 2 World Championships

#68

It is so weird reading all these comments. Almost half of them start out as "it's impressive.. but". Is it human nature or are HN commenters just so sceptical/negative of all the new tech? One of the OpenAI guys mentioned that they could potentially use the same technique in real life applications like surgery. Surgery is not "just" run on a computer it has a physical component too. Is that really the next step or wa…

HN culture strongly discourages unsubstantive comments like "wow, this is great" that don't add to the conversation. It's hard to comment positively and substantively on a story like this, because it requires you to understand the subject matter well enough to describe some interesting feature, or bring up related work. It's usually much easier to find a flaw, or something that looks like a flaw to fellow non-experts.

This comment section actually seems ok, but usually you'll see mostly neutral or negative comments.

Re: OpenAI at the Dota 2 World Championships

#69
post #43

Earlier quoted context omitted.

Raw screen is not as big of a deal as it would seem, in DotA and Starcraft you can glimpse most of the information you can see on a screen through the minimap.

It would seem to me that this is an oversimplification. Doing screen based video processing is not simple and is not exact especially where camera movements are independent of game dynamics as well as the difficulties inherent in doing fast frame classification- there has been some good work in this direction recently but far from perfect(yolo, ssd). The route deep mind took in partnering with the game manufacturers…

Oh I misread the "Direct access to game API section", I thought they meant making it so that the machine can only see "what's on their screen", so if someone were to cast a long range skill for instance they wouldn't get that information. Which is a distinct advantage for machines.

That is indeed, by no means trivial. I do think it's pointless though, that's just another exercise for the sake of making it another exercise.

Re: OpenAI at the Dota 2 World Championships

#70
post #25
post #18

Earlier quoted context omitted.

To be fair, the 1v1 Shadow Fiend matchup with these rules is A LOT about exact distances and timings. To which the bot has direct API access and the human has to got the whole GPU -> screen -> eyes -> brain -> muscles -> mouse -> CPU cycle.

Agreed, bot will always be good at spell casting, last hitting, which actually is a major factor in winning(the sooner you get higher level you can zone the other player).

Especially for Shadow Fiend, which, for those who don't know, gets extra damage for every last-hit. If you're good at it, you can have extra damage equating to a late game item as an early game character.
Post reply on HN