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

DeepMind Challenges for StarCraft

gamasutra.com

Post reply on HN
Print view View original

101–110 of 137 posts

Re: DeepMind Challenges for StarCraft

#101

Post by ygra »

timwaagh wrote: post #93

I'm rather suprised there are no player beating ais for starcraft. i dont think it should be very hard as decisions need to be made very quickly in this game and humans are bound to make more than a few mistakes. but perhaps i'm underestimating the problem. it may have something to do with the limitations of the api though.

You're underestimating the problem. And there are several:

- Controlling individual units: scouting, resource gathering, micro-management while attacking/retreating, ...

- Resource management and planning: knowing when and where to expand, save resources for certain points in the tech tree, e.g. so you can immediately build a bunch of units once a crucial research or building is complete, ...

- Figuring out what the opponent is building and what he could be doing: where is he and what race, what tech buildings are there, when does he start to harvest gas, key tech buildings may be hidden in other places of the map, essentially even not being attacked with certain units at a certain time may be a clue to some strategy

- Reading and using the terrain properly: High ground has an advantage, Terran can wall off certain entrances with a depot and a barracks, where would be good places for drops (and with what units), figuring out where the enemy could come from and which places to defend – all those vary by map quite a bit

- Finding a good strategy to counter what the opponent is doing: This ties into a few of the things above, because first it must be known what the enemy is doing. Keeping the strategy flexible enough to counter the enemy's counters, etc.

Some of those are necessary to solve in real-time (unit movement), some of those can be run in background and don't need to be frame-exact (strategy, etc.), which helps a bit. But there are still quite a few parts to playing the game, all of which are interconnected and require information from other parts. Many successful bots currently are rush bots, which exploit that most bots are weak in the beginning and are easy to create. Others rely on a number of fixed strategies and build orders that are chosen based on what the enemy is observed to do.

Re: DeepMind Challenges for StarCraft

#102

Post by daveguy »

egeozcan wrote: post #97
daveguy wrote: post #96

Earlier quoted context omitted.

Raw pixels. And the score. The score was separate, or rather a signal representing increased score. Relevant quote from the paper: "The emulator’s internal state is not observed by the agent; instead it observes an image xt ∈ Rd from the emulator, which is a vector of raw pixel values representing the current screen. In addition it receives a reward rt representing the change in game score." The paper: http://arxiv.o…

It would be great if it could also learn to tell if it was doing fine.

Agreed. I think that is one of the major drawbacks/limitation/unaddressed aspects of deep learning algorithms -- they are primarily supervised learning. Supervised in the sense that you have to explicitly identify good and bad examples. Determination of what is good and bad itself (figuring out that number at the upper right side of the screen is a score) would be a major breakthrough with implications far beyond game playing. DNN has been a breakthrough with much better accuracy and discrimination capabilities of a complex neural network. It still requires that the researcher point out what is good and bad. We still need a just-as-significant breakthrough in unsupervised learning.

Re: DeepMind Challenges for StarCraft

#103

Post by shultays »

scottjad wrote: post #41

Starcraft is a game that is played with mouse and keyboard (or game controller on console). I think in order to say an AI has beat a human, it should have to use the same interface that a human uses. That's part of the definition of the game. In Go, anyone can place a stone as well as anyone else, it's deciding where to place a stone that's the game. In Starcraft, not everyone can control the interface to the game eq…

AI is still giving same commands as a player, move unit to there, attack that unit. Even if it was limited to do one action per frame, assuming 60 fps, that would be 3600 action per minute and still x15-x20 times of a good sc player.

Further limiting the apm would be no different than limiting the AI itself. It wouldn't be any different than limiting the cpu/memory usage imo.

Re: DeepMind Challenges for StarCraft

#104

Post by martinald »

derefr wrote: post #69
barbs wrote: post #52

Earlier quoted context omitted.

From the 2015 AIIDE Starcraft AI competition report[0]. > ""Why not StarCraft 2?" > This is the question we always get asked when we tell people we are doing a BroodWar AI competition. This competition relies completely on BWAPI as a programming interface to BroodWar. BWAPI was created by reverse engineering BroodWar and relies on reading and writing to the program memory space of BroodWar in order to read data and i…

It has always seemed to me that there is a truly "proper" way to construct an AI 'bot' for a networked game, and that is to write your own headless game client software that connects to the game's network—emulating the physics well enough to avoid desynchronization with one's opponent, but otherwise being entirely its own program, rather than having anything to do with the reference client per se .

It's really really hard even for developers of the actual client to avoid desyncs. Trying to reverse engineer the exact physics engine down to the bit would be a massive achievement technically in itself, I'd say.

Re: DeepMind Challenges for StarCraft

#106

Post by sesutton »

DanBlake wrote: post #3

One of the issues to overcome is the advantage machines can have in individual unit manipulations vs what a human can do in terms of APM https://www.youtube.com/watch?v=SZawUuAccnc

I'm a little worried that the AI will quickly develop a simple early game strategy where it sacrifices economy for early units (an "all-in") that never loses because of its perfect unit control.

The AIs unit control speed may have to be limited to get interesting results where it has to use strategy.

Re: DeepMind Challenges for StarCraft

#107

Post by smilekzs »

dogma1138 wrote: post #24
smilekzs wrote: post #12

Earlier quoted context omitted.

The game is capable of accepting a much higher APM than Korean pro players are capable of pulling off even at peak. IMHO the AI should be subject to an APM limit (e.g. sliding window).

Not if it's done over the normal channels (USB) many players not even "pro korean" ones can easily encounter input issues like key rollover limits and polling rates on your mouse. USB has quite a bit of limitation for HID's and allot of the gaming mice and keyboards use various hacks to overcome them. Games and OS also have limits I can easily issue more commands using my keyboard than SC2 can handle so not all key p…

IIRC the entire input polling/NKRO issue can be solved by simply switching from USB2.0 FS (aka USB1.1) to HS, which requires a more powerful keyboard/mouse controller.

Re: DeepMind Challenges for StarCraft

#109

Post by JabavuAdams »

scottjad wrote: post #41

Starcraft is a game that is played with mouse and keyboard (or game controller on console). I think in order to say an AI has beat a human, it should have to use the same interface that a human uses. That's part of the definition of the game. In Go, anyone can place a stone as well as anyone else, it's deciding where to place a stone that's the game. In Starcraft, not everyone can control the interface to the game eq…

> Starcraft is a game that is played with mouse and keyboard (or game controller on console). I think in order to say an AI has beat a human, it should have to use the same interface that a human uses. That's part of the definition of the game.

That's the easy part, though. If you can make a SC AI play well against a human you can certainly send mouse events into a window. It really doesn't show anything.

See for example Sikuli, which you can use for automated testing.

Post reply on HN