Good AI is Critical for the Next Big Step in Civ Games - Page 2

Mark_Everson · January 23, 2003, 11:17

Quote:

Originally posted by Lemmy
Adaptive AI wouldn't require much processor power, it's basically looking things up, and saving new things. It would require more space, but the space needed can be brought down by an efficient design, for example by not saving all cases, but only the most "important" ones, and keep averages for the remaining cases.

Hi Lemmy,

There is a way to do basically what you're looking for, in a practial sense. But it requires human intervention. Essentially players can provide input into the expert system. All that is required is that the game producer provide time after a large player base is available so that the programmer can update the expert system based on players' experiences and ideas. Of course the expert system needs to be flexible enough to represent the player strategies accurately in terms of their realms of applicability. That is a very tall order already!

Its nowhere near as sexy in an AI sense as what you want to do, but it will work with current technology for a game with the depth of civ

Panag · January 27, 2003, 22:04

hi ,

, Soren has not reached his limits yet , .......

the real Q is , a better AI shall need a better comp , .....

do most people have comps that require what is needed for an improved AI , ....

have a nice day

Leland · January 28, 2003, 01:36

I don't think that computing power is that relevant in TBS genre... serious strategy gamers would gladly wait for 10-20 minutes per turn if they got more intelligent AI in exchange. The fact that games generally don't have an option of setting AI "depth" in this manner seems to imply that currently AIs are based on clever heuristics rather than brute force.

Panag · January 28, 2003, 08:49

Quote:

Originally posted by Leland
I don't think that computing power is that relevant in TBS genre... serious strategy gamers would gladly wait for 10-20 minutes per turn if they got more intelligent AI in exchange. The fact that games generally don't have an option of setting AI "depth" in this manner seems to imply that currently AIs are based on clever heuristics rather than brute force.

hi ,

well aslong as they include an option ; "turns in 1 min" - "turns in 20 min" ( with extra intelligent AI ) , ......

no thanks , no-one is going to wait 20 minutes , .........

3 - 4 tops , thats it , it has to be possible , no more then that and a more intelligent AI

have a nice day

Mark_Everson · January 29, 2003, 22:20

I don't want the turns to take any more than about 20s to crank! Nearly instantaneous is best.

The best thing to do for AI is run it while the player is moving! Threading immensely changes what you can accomplish with the AI. GalCiv has, and Clash will have threaded AI, and I think its going to change the playing field for AI in the genre. But then again, I'm just a little bit biased.

Panag · January 30, 2003, 06:43

Quote:

Originally posted by Mark_Everson
I don't want the turns to take any more than about 20s to crank! Nearly instantaneous is best.

The best thing to do for AI is run it while the player is moving! Threading immensely changes what you can accomplish with the AI. GalCiv has, and Clash will have threaded AI, and I think its going to change the playing field for AI in the genre. But then again, I'm just a little bit biased.

hi ,

what is really needed is a state of the art AI that can run turnless in real time (!) , ......

but , is this possible

have a nice day

Mark_Everson · January 31, 2003, 09:53

Quote:

Originally posted by panag
what is really needed is a state of the art AI that can run turnless in real time (!) , ......

but , is this possible

have a nice day

Hi panag, I don't get what you mean. Civ isn't real-time and I hope it never is! But as I said above, thinking in the background while the player moves is vital for good AI.

Panag · January 31, 2003, 11:06

Quote:

Originally posted by Mark_Everson

Hi panag, I don't get what you mean. Civ isn't real-time and I hope it never is! But as I said above, thinking in the background while the player moves is vital for good AI.

hi ,

well there are seveal games around who have something that comes a bit close , but even they are far from being state of the art , .....

remember C&C , tiberian sun , firestorm, ....

something like that , but in real time , where there is a constant AI who runs , .....

sincve some people shall not like this , this could be solved with a basic "on / off" , ....

have a nice day

Andemagne · February 2, 2003, 09:09

yes it is. are games with not so good AI worth playing?
(=> are Civ2 & Civ3 worth playing 'cos they have so bad AI?)

MrBaggins · February 2, 2003, 22:50

Quote:

Originally posted by Mark_Everson

Hi panag, I don't get what you mean. Civ isn't real-time and I hope it never is! But as I said above, thinking in the background while the player moves is vital for good AI.

Intelligent human players will then play 'speedciv' to keep the AI from utilizing down time to think.

Panag · February 3, 2003, 03:40

Quote:

Originally posted by MrBaggins

Intelligent human players will then play 'speedciv' to keep the AI from utilizing down time to think.

hi ,

true , most players like to think in advance , after a couple games they shall have outlined the big lines they want , ......

with a bit of strategy its easy to beat the AI , so after a couple times its fun , after 20 times its no longer fun , it just becomes past-time , ....

the AI in civ three - PTW needs to be improved , ......

have a nice AI day

Urban Ranger · February 11, 2003, 03:10

Quote:

Originally posted by Mark_Everson
4. At this point, if it's required that we stop, we just pick from the best indicated strategies. If there is extra time, there is a large variety of things that can be done. The one that I think is most attractive, is to encode the strategies as as individuals in a population of a genetic algorithm approach. We then pursue more simulations as in 3, but while tweaking the strategies through mutation and crossover. If the rule-based approach can give us any hints as to which are the most productive things to change, the more the better. The chromosomes will not generally contain numbers at all, but contain strategic objects. "Take Berlin", "obtain alliance against the Greeks". A mutation in the strategy might be something like "it would be nice to get an alliance against the Greeks, but it isn't that essential", or "it's Imperative". I'm not sufficiently foolish to think that this approach will very frequently give an answer better than the rule-based approach. However, it has the potential to break the AI out of the rut of always doing "predictable" things.

You certainly can play around with evolutionary algorithms, but there are certain problems with that. First of all, your AI needs to be completely flushed out. Then, you need to somehow induce mutations and put them through a set of evaluation criteria, neither of which is a simple task. Another thing is you need to run the evolutionary algorithms as a meta-game, i.e., you need to do it outside the game, you can't do it when a game is running.

My take on this is heuristics. Start with a decent expert system, but add the capacity to learn from human players. That way, the AI is never obsolete. It can learn from playing with humans, and the developers can keep training it even after the game is released, and release the new code and/or database to make the AI smarter. This way, the game can be challenging for a long time.

Lemmy · February 12, 2003, 18:46

Quote:

My take on this is heuristics. Start with a decent expert system, but add the capacity to learn from human players. That way, the AI is never obsolete. It can learn from playing with humans, and the developers can keep training it even after the game is released, and release the new code and/or database to make the AI smarter. This way, the game can be challenging for a long time.

I believe this is what the GalCiv dev team has planned for their own AI/

Mark_Everson · February 12, 2003, 20:22

Hi UR:

Quote:

Originally posted by Urban Ranger
You certainly can play around with evolutionary algorithms, but there are certain problems with that. First of all, your AI needs to be completely flushed out.

I don't know what you mean by this. Care to elaborate?

Quote:

Then, you need to somehow induce mutations and put them through a set of evaluation criteria, neither of which is a simple task.

My ideas on this are crudely described in the AI documentation. There will clearly be some art involved in this, but I don't think it'll be especially difficult.

Quote:

Another thing is you need to run the evolutionary algorithms as a meta-game, i.e., you need to do it outside the game, you can't do it when a game is running.

What is this statement based on? AFAIK it is utterly incorrect

. We are going to run the AI for Clash in-game in one or more separate threads.

The GalCiv approach, which appears to be quite good is, as Lemmy says, like what you appear to be describing. I expect it'll work quite well, but takes more dedication than the usual game software company is willing to expend. I'm looking forward to GalCiv rather a lot

.

Urban Ranger · February 12, 2003, 23:09

Quote:

Originally posted by Mark_Everson
I don't know what you mean by this. Care to elaborate?

That means the AI has to cover all the possibilities. If something is not in the code/database, it doesn't know how to handle it, and goes haywire in all likelihood. But knowing what to expect in the game is directly related to experience, which isn't something a programmer will have when writing the AI routines. You can tell this by all the expert systems in all the 4X games, they are okay, but ultimately they all rely on cheating - breaking the rules.

Quote:

Originally posted by Mark_Everson
What is this statement based on? AFAIK it is utterly incorrect

. We are going to run the AI for Clash in-game in one or more separate threads.

Based on the genetic algorithms I have seen

Not a whole lot, mind you, but the point is to have the individuals and an environment. If the individuals are the AI routines, the environment will be an arena with a set of criteria where these routines compete.

I am not 100% sure you cannot run the arena in situ, but you will need to spawn hundreds if not thousands of individuals for thousands of generations. This will be much better handled by a meta-game, i.e., run the arena outside the game itself to do the evolution.

A drawback of any evolutionary algorithm is that a solution is "better" only in comparison to other, presently known solutions; such an algorithm actually has no concept of an "optimal solution," or any way to test whether a solution is optimal. This is not entire true of Civ.

Mark_Everson · February 14, 2003, 08:13

Hi Urban Ranger, thanks for the detailed response.

Quote:

Originally posted by Urban Ranger
That means the AI has to cover all the possibilities. If something is not in the code/database, it doesn't know how to handle it, and goes haywire in all likelihood. But knowing what to expect in the game is directly related to experience, which isn't something a programmer will have when writing the AI routines. You can tell this by all the expert systems in all the 4X games, they are okay, but ultimately they all rely on cheating - breaking the rules.

What you say is a problem with an AI constructed by very simple heuristics of the if/then type. I agree that that is a standard for AI now, but that is one of the things I think needs to be discarded to get the next step in AI for civ games. We will certainly have a set of heuristics to suggest good starting strategies, but where we differ from the standard approach is that we then passed several strategies to see which fits better in a given situation. What the Clash AI does is it has a cartoon of the world for each type of strategy any uses the cartoon version, which is based on game parameters, to estimate success of a strategy. Each type of strategy, FE "attack area" to try to take control a given area of the map, has its own cartoon of the world that outlines the big issues involved. It's kind of the way people think about it. This is gone over in great detail in our "AI Plan" document in "AI -- the thread" on the Clash forum. There is a link to that thread in the third post of this thread.

Quote:

Based on the genetic algorithms I have seen

Not a whole lot, mind you, but the point is to have the individuals and an environment. If the individuals are the AI routines, the environment will be an arena with a set of criteria where these routines compete.

I am not 100% sure you cannot run the arena in situ, but you will need to spawn hundreds if not thousands of individuals for thousands of generations. This will be much better handled by a meta-game, i.e., run the arena outside the game itself to do the evolution.

Well, it's back to the books for you!

That is in fact a typical way GAs are run for engineering optimizations and other things warrior doing your very best to find the absolute global optimal. This is a GAME and I certainly don't attempt to optimize all my strategies in it, why the heck should the AI? The job of GAs in Clash is to just take stabs in the dark looking for "surprise" approaches, that are different from Clash-standard AI, to make the AI less predictable.

Quote:

A drawback of any evolutionary algorithm is that a solution is "better" only in comparison to other, presently known solutions; such an algorithm actually has no concept of an "optimal solution," or any way to test whether a solution is optimal. This is not entire true of Civ.

When you can play an optimal game of civ I will worry about it. We are only looking for an entertaining and sometimes surprising AI.

Urban Ranger · March 5, 2003, 03:40

Just read something interesting about Galatic Civilization. Brad said they got 6 AI routines, 5 for the major powers and 1 for the minor powers. Strangely, the 5 are the same thing written by different programmers. They did it to prevent a player from exploiting weaknesses in any one of the routines.

I reckon this can be eliminated if the AI can learn from the human players. In this way, programmers can train the AI so it can take on various approaches (aggresive, defensive, etc.) without extra programming efforts.

A very good example is the old, old Atari game called Castle. Not sure if you can still find it anymore, though.

mactbone · March 5, 2003, 22:00

Speaking of adaptive ai etc.

Ever hear of EvoChess? I discovered this last year or so and started running it. It's basically a simple chess program that uses "genetic" models to get the best AI. Not only does it pit your generated algorithms against already made "good" algorithms - there is also some migration over the 'net. This was all part of someone's research paper.

I can't find a simple page with everything so here are the various pages (it helps if you know German, I'm sure but most of the documentation is in English)
http://ls11-www.cs.uni-dortmund.de/p...h/evoChess.htm
http://www.science-at-home.de/index1.htm?/evochess.htm (only in German)

Ahhh, here's probably the best info.
http://www.aspenleaf.com/distributed...ib-recent.html
Scroll down to Qoopy, about halfway down.

Panag · March 6, 2003, 06:04

hi ,

after been forgotten for a long time ; http://www-cs-students.stanford.edu/~soren/

its Sorens work on AI , .....

and he is good , ....

have a nice day

Mark_Everson · March 7, 2003, 09:37

Quote:

Originally posted by Urban Ranger
Just read something interesting about Galatic Civilization. Brad said they got 6 AI routines, 5 for the major powers and 1 for the minor powers. Strangely, the 5 are the same thing written by different programmers. They did it to prevent a player from exploiting weaknesses in any one of the routines.

I reckon this can be eliminated if the AI can learn from the human players. In this way, programmers can train the AI so it can take on various approaches (aggresive, defensive, etc.) without extra programming efforts.

Hi UR:

I'm no sure I get it. What exact mechanism do you forsee for the AI learning from players? Is it in-game or outside the game (FE by programmers adding new player strategies by hand) that this is done?

Thanks for the references panag and MacTBone. I'll try and check them out at some point.

LDiCesare · March 9, 2003, 17:20

Soren's AI work is a very clear indication of what he did in Civ3: He says the game must have a small anough search space that the computer can search it efficiently, which means dumbing down game mechanisms so that whatever the ai doesn't know how to exploit effectively should be removed. Someone wrote an article on Civ3 showing that was exactly what happened.

On a different game, MOO3 seems to have gotten out with a pitiful enemy AI (the viceroys seem to be good but not always doing what you expect them to do). The interesting bit of info is that the MOO3 team changed AI settings 2 weeks prior to release because it was too hard, and ended up with a too easy AI. Since that looks like algorithmic rather than text files tweaking, it is quite sad. On another side, GalCiv is supposed to have an efficient AI which can be tweaked by changing just one figure, which decides whether or not it will recognize threats and decide of certain advanced tactics. I'd really like to see this one come out in France soon...