Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

StarCraft Micromanagement with Reinforcement Learning and Curriculum Transfer

arxiv.org

Post reply on HN
Print view View original

21–30 of 54 posts

Re: StarCraft Micromanagement with Reinforcement Learning and Curriculum Transfer

#21

Post by taurath »

bottled_poe wrote: post #7
axxl wrote: post #5

Earlier quoted context omitted.

There is some stuff like this, you would see it on the ladder in SC2. Hacks that would perfectly split your marines against incoming banelings (they explode and do splash damage so the more you split the less effective they were), or auto burrowing/unburrowing roaches (which hides them from the enemy and they heal underground). It sounds cool but it’s way cooler when you saw a pro doing all that stuff themselves almo…

Check this out. 100 zerglings vs 20 siege tanks - https://www.youtube.com/watch?v=IKVFZ28ybQs

Interesting. It seems that this is possible mostly by the lag time between target acquisition and firing by the siege tanks, compounded with the relatively high speed of the zerglings - most any other unit in the game wouldn't be able to react in time.

Re: StarCraft Micromanagement with Reinforcement Learning and Curriculum Transfer

#22

Post by justicezyx »

s_m_t wrote: post #9
calebh wrote: post #6

For those who don't know Starcraft: microing refers to the small scale/precise maneuvering of units during a battle, as well as activating special unit abilities at certain times. In the lower level leagues microing is not as important as having a good overall large-scale "macro" strategy. The microing really starts coming into play at the higher levels where it can give the edge needed to win a battle. I am not very…

How can this be so? In my experience playing Starcraft the decision space for macro is very, very, very limited. You are essentially locked into a handful of viable build orders for the early game, a handful of timing attack opportunities in the early and mid game, and after that things veer so off track that very few players do more than wing it.

I am sure AI friends can come up with some interesting strategies.

Re: StarCraft Micromanagement with Reinforcement Learning and Curriculum Transfer

#23

Post by s_m_t »

tialaramex wrote: post #20
s_m_t wrote: post #9

Earlier quoted context omitted.

How can this be so? In my experience playing Starcraft the decision space for macro is very, very, very limited. You are essentially locked into a handful of viable build orders for the early game, a handful of timing attack opportunities in the early and mid game, and after that things veer so off track that very few players do more than wing it.

This is sometimes called the game's "meta" but it's very unclear how much of this is real and how much is basically just fashion. Some good but not quite star quality players like Ketroc deliberately ignore the meta. Ketroc plays with loads of Ravens, he did this when people said Ravens were strong but hard to use, he still did it when the accepted wisdom was that they were now useless, and last time I looked he was…

Sorry, a bit of ambiguity here, I was talking about SC:BW, as the paper is also concerned.

I think that SC2 and BW differ in this respect because SC2 has been constantly updated throughout its competitions to make more strategies and units viable. The last balance patch for BW was two years after release in 2001 (although a recent bug fix is said to have slightly buffed valkryies, there is some dispute about whether this actually pans out in practice but I haven't looked into it).

Re: StarCraft Micromanagement with Reinforcement Learning and Curriculum Transfer

#24

Post by daemonk »

axxl wrote: post #5
daemonk wrote: post #4

I want an AI assisted mode with API hooks where people can code their own sophisticated "macros" that can be activated with a command/button. Limit the code to X number of characters. It would be awesome to see a tournament with this.

There is some stuff like this, you would see it on the ladder in SC2. Hacks that would perfectly split your marines against incoming banelings (they explode and do splash damage so the more you split the less effective they were), or auto burrowing/unburrowing roaches (which hides them from the enemy and they heal underground). It sounds cool but it’s way cooler when you saw a pro doing all that stuff themselves almo…

I appreciate the manual dexterity, but imagine the amount of strategy that AI assisted play would open up.

Re: StarCraft Micromanagement with Reinforcement Learning and Curriculum Transfer

#25

Post by fosco »

daemonk wrote: post #4

I want an AI assisted mode with API hooks where people can code their own sophisticated "macros" that can be activated with a command/button. Limit the code to X number of characters. It would be awesome to see a tournament with this.

if this were available in true form, would that reduce the game to the fastest unit no matter what the damage.

as an example see the zerg vs tank below.

I understand the idea of code manipulation but what might be better is just having human programmed AI's fight each other (which I am fairly sure is going and can be viewed live but I do not have the link handy.

Re: StarCraft Micromanagement with Reinforcement Learning and Curriculum Transfer

#26

Post by vkou »

s_m_t wrote: post #9
calebh wrote: post #6

For those who don't know Starcraft: microing refers to the small scale/precise maneuvering of units during a battle, as well as activating special unit abilities at certain times. In the lower level leagues microing is not as important as having a good overall large-scale "macro" strategy. The microing really starts coming into play at the higher levels where it can give the edge needed to win a battle. I am not very…

How can this be so? In my experience playing Starcraft the decision space for macro is very, very, very limited. You are essentially locked into a handful of viable build orders for the early game, a handful of timing attack opportunities in the early and mid game, and after that things veer so off track that very few players do more than wing it.

The decision space for macro is very limited, but correct execution of it extremely important.

The basics (Spend your money, don't queue units, don't get supply blocked) are trivial. Executing them perfectly for 20 minutes, during a stressful game, is impossible. The best players in the world can't do it.

Re: StarCraft Micromanagement with Reinforcement Learning and Curriculum Transfer

#27

Post by d1zzy »

daemonk wrote: post #24
axxl wrote: post #5

Earlier quoted context omitted.

There is some stuff like this, you would see it on the ladder in SC2. Hacks that would perfectly split your marines against incoming banelings (they explode and do splash damage so the more you split the less effective they were), or auto burrowing/unburrowing roaches (which hides them from the enemy and they heal underground). It sounds cool but it’s way cooler when you saw a pro doing all that stuff themselves almo…

I appreciate the manual dexterity, but imagine the amount of strategy that AI assisted play would open up.

If you're looking for more strategic gameplay instead of micromanagement, Starcraft isn't the best game to pick. Lowering the importance of high APM for a game like Starcraft through the use of smart macros is just a hack, why not just play (or design) a game that doesn't need any of that. Of course, the obvious choice is any turn based game.

But even if you want to stay in RTS land, one of the more recent titles that comes to mind where micromanagement is not as important and strategy/planning is very important is Ashes of the Singularity. Spoiler: the game may seem "slow" compared to Starcraft but that's also part of reducing the importance of high APM.

Re: StarCraft Micromanagement with Reinforcement Learning and Curriculum Transfer

#28

Post by acomar »

loser777 wrote: post #18
taurath wrote: post #17

Earlier quoted context omitted.

Just like playing against MarineKing would get stale quickly for most players :P The problem with starcraft is that its like chess - for a lot of players it really DOES get stale pretty quickly, with only one or two interesting moments in a game (assuming no early blunders). The games that are interesting involve near perfect play for 10-20 minutes and then frequently one split second decision. Macro changes or new s…

The difference is that an interesting AI could likely be scaled back meaningfully for lower skill levels. Make MarineKing play with only the mouse against a lower skilled opponent? The results might be less stale. Rate limiting a brute-force AI would make it trivially bad.

Not really. Stopping MarineKing from sending tons of APM is only part of the challenge. His choices for what to do in any given frame (yes, frame) are also just better than basically every player on the ladder. If you remove his keyboard, you'll find that he still wins most games against folks who don't know the game.

Re: StarCraft Micromanagement with Reinforcement Learning and Curriculum Transfer

#29

Post by acomar »

s_m_t wrote: post #9
calebh wrote: post #6

For those who don't know Starcraft: microing refers to the small scale/precise maneuvering of units during a battle, as well as activating special unit abilities at certain times. In the lower level leagues microing is not as important as having a good overall large-scale "macro" strategy. The microing really starts coming into play at the higher levels where it can give the edge needed to win a battle. I am not very…

How can this be so? In my experience playing Starcraft the decision space for macro is very, very, very limited. You are essentially locked into a handful of viable build orders for the early game, a handful of timing attack opportunities in the early and mid game, and after that things veer so off track that very few players do more than wing it.

It's really not. It seems limited because of what people have decided are strategies that win and so the strategies that people try. However, this is not, entirely, what macro refers to. Try playing a game of starcraft. Try to keep up with worker construction, base management, unit construction, and basic map presence. It's harder than you think, even at the top levels of the game.

Re: StarCraft Micromanagement with Reinforcement Learning and Curriculum Transfer

#30

Post by WhompingWindows »

Benjammer wrote: post #10
calebh wrote: post #6

For those who don't know Starcraft: microing refers to the small scale/precise maneuvering of units during a battle, as well as activating special unit abilities at certain times. In the lower level leagues microing is not as important as having a good overall large-scale "macro" strategy. The microing really starts coming into play at the higher levels where it can give the edge needed to win a battle. I am not very…

I actually find the advanced micro to be far more interesting in the edge cases where you can break the game's balance system. I remember when the Brood War api first was released a while ago, there was an AI bot that won some competition with just mutalisks because the AI micro prevented them from overlapping their AOE hit boxes as they moved around. So you had mass mutalisks that the opponent couldn't use AOE again…

You don't even need an AI to do this. ByuN micros reapers so well that he was literally the only player at the top echelons of play who revealed they could be OP. Blizzard nerfed the reaper twice specifically due to his play.
Post reply on HN