Emergent AI In an Indie RTS Game 146
x4000 writes "My recent RTS game uses a new style of AI that hybridizes rules-based AI with emergent AI logic. As a disclaimer, I'm really not an AI programmer at all — my background is in databases, financial modeling, etc. But it just so happens that database experience, which often involved distilling data points from multiple sources and then combining them into suggested decisions for executives, also makes a great foundation for certain styles of AI. The approach I came up with leans heavily on my database background, and what concepts I am familiar with from reading a bit about AI theory (emergent behavior, fuzzy logic, etc). The results are startlingly good. Total development time on the AI was less than 3 months, and its use of tactics is some of the best in the RTS genre. I'm very open to talking about anything and everything to do with the design I used, as I think it's a viable new approach to AI to explore in games, and I'd like to see other developers potentially carry it even further."
Article summary: (Score:5, Insightful)
"I just wrote a game. Heres some techy stuff to get it posted on slashdot. Yay advertising!"
Re: (Score:2)
Thanks for the Culture of Cynicism comment -- we definitely need more people like you around that needlessly shit on others doing new stuff.
Re:Article summary: (Score:4, Insightful)
It is very light on implementation details.
Which is unfortunate, because it's a really interesting topic.
Basically what he says is that he has three levels of AI: strategic, sub-commander and unit. Units look out for themselves, strategic is rule based and sets strategic goals, and sub-commander is the fun part: it makes units work together through flocking.
That's pretty much it. Very interesting, but a bit more meat would have been nice.
The slashvertisement worked, though. I'm very tempted to buy this game to see if the AI is really good, even though I hate RTS games.
Re: (Score:3, Interesting)
I agree that this is very interesting, but the sheer asymmetry of the setup (players begin with four ships versus 20,000+ for the AI) means that we don't really have a chance to compare the AI to human intelligence. The concepts seem sound, but they need to be implemented in combination with an economic model which will allow symmetric human v. AI showdowns. Something like this would be very cool, for example, in the next iteration of the Civilization series.
To his credit, the author acknowledges these limi
Re: (Score:2)
I agree that this is very interesting, but the sheer asymmetry of the setup (players begin with four ships versus 20,000+ for the AI) means that we don't really have a chance to compare the AI to human intelligence.
It's an RTS, so it was never going to give an accurate measure of strategic intelligence anyway. But on the whole, I agree with your point. I want to see this in action in a good, level playing-field TBS game.
Re:Article summary: (Score:4, Interesting)
It's a great approach. It's one of those quasi-obvious AI schemes which people have been kicking around for years but typically haven't wanted to expend the CPU cycles on in major projects. I remember an article around 1997 wherein they asked game development studios to explain how games were going to change in the next couple of decades, and one of the chief topics that kept coming up was more natural AI (less rigidly rule-based) which could learn by observing outcomes. Modelling AI on military organization is fairly clever for the same reason that military organization (at least in battle) is clever - the individual soldier needs little autonomy or smarts, just so long as everybody further up the chain from them is at least incrementally smarter, if less capable in direct conflict.
The most notable use of a system like this in recent years has to be the Director in Left 4 Dead, which is an overarching control scheme which issues orders to the game world and the units in it. These units are free to do what they do best - navigate around the map and eat players without having to worry about strategy. The Director is free to select the paths by which the units will arrive without having to waste significant time on fine pathfinding, as the units themselves will do the bulk of the work.
I look forward to more games with heirarchical AI. I've written some basic tests of this exact sort before (basic "burds"-looking stuff) and the resulting behavior is extremely intuitive and ordered-looking.
Re: (Score:2)
I've written some basic tests of this exact sort before (basic "burds"-looking stuff) and the resulting behavior is extremely intuitive and ordered-looking.
But can it kick my ass in a fair fight? That's what I want to know.
Re: (Score:2)
He should sell his technology to Valve. Last night in L4D I watched two NPC teammates casually walk into fire and burn to death. I think their AI could use some work.
player mimicry (Score:5, Interesting)
could you catalog user actions and use them as possible inputs for your AI?
like this: http://slashdot.org/comments.pl?sid=1165583&cid=27243769 [slashdot.org]
it would seem this might make for the most challenging AI - one that
learns and mimics good human players. I have yet to see any games
that can do this well
Re: (Score:1)
Re: (Score:2)
I think the Mimiry suggestion is more comparable to a chess player who has played a lot of games, studied other people's games and studied books on opening theory and that sort of thing. That gives a good player a big knowledge base of good moves and a feel for how the results of those moves might turn out. A self-learning system would expand its base of good moves with each new game it plays.
Re: (Score:2)
Unfortunately, you are giving chess a little too much credit here. An unbeatable chess strategy really is to have an internal catalog of board states and optimal moves from those states.
For a less rigid game a strategy of recording player moves becomes intriguing, but also extraordinarily more complex in that you must catalog sequences of related actions. AI by macro?
Re: (Score:2)
Unfortunately, you are giving chess a little too much credit here. An unbeatable chess strategy really is to have an internal catalog of board states and optimal moves from those states.
No, an unbeatable chess strategy is calculating the entire game to the end (or at least to an obviously won osition). I'm just explaining how chess strategy works in practice. How am I giving chess too much credit?
Re: (Score:1)
I realize that players at that level probably mostly play from a memory of board states, but there are so many possibilities that knowing which ones to focus on let's you analyze the possibilities that are most likely. The unbeatable chess strategy you describe is for a computer with a ton of memory and a fast search capability, not for humans or human-like AI.
Back to
Summary useless (Score:5, Informative)
Re: (Score:2, Insightful)
Even I, 12 years ago at the University, had to implement a miserable little game where the A.I. was at a lower level than an all seeing, all knowing AI that makes decisions, each agent with it's set of "intelligent", adaptive rules.
So... how is this concept new (or NEWSworthy, for that matter)?
Re: (Score:2)
So... how is this concept new (or NEWSworthy, for that matter)?
This is games.slashdot, and good AI still hasn't been done in games. If this results in some decent AI, it's worthy of slashdot. But a bit more data on either the quality of the AI or the technical side of it would have been nice.
Re: (Score:2)
how is this concept new (or NEWSworthy)
By linking sufficient AI together in a functional framework which could become a commercially and financially viable product. This is a far more difficult task than a single project at University which does not have to appeal to a wide audience with often conflicting requirements.
Re:Summary useless (Score:5, Informative)
Re:Summary useless (Score:5, Interesting)
In short, it segregates the controller AI from the unit AI.
I can remember when playing SupCom against easy opponents, if I didn't build any defenses, the opponent would stop sending units into my base for a while, instead moving them around in circles somewhere outside it. Even an easy/stupid AI has to make decisions that a stupid person would make. Mainly - CHARGE!
But at the same time, the best AIs were omniscient. They'd target your first fusion reactor with bombers as soon as you build it... It became so predictable that I'd build a shield, turn it off, build a fusion, then turn the shield back on.
But in SupCom there were all the modded AIs which chose more interesting tactics, and you could enable cheating for any AI to give them a resource and difficulty boost.
I think to be realistic, AI has to be segregated even more. You need an Advisor, which decides the direction things should be going. Do we need more resources? Do we have an urgent matter to deal with, like artillery pelting the base? The Advisor should just decide what to tackle, and then pass that on to the other AI subsystems. Like a player, the AI subsystems should take time to do stuff. An AI can't instantly decide what's an entrance to his base. He has to study where the units usually come from, and locate possible landing zones for air transports, etc.
To ramp up the intelligence, you let it assign more issues to its subordinates more rapidly, and you let it learn across missions. Pretty soon the AI can tackle any frequently played map.
Re: (Score:3, Interesting)
These games use fairly simple monster AI that usually follows simple rules, so it may be much simpler than what this guy claims. However I think it must be applied per creature. Some types of monster will run and cower when y
Re: (Score:3, Informative)
Diablo series uses heuristics, pure and simple. Each unit has a set behaviour rule, with no adaptive whatsoever.
Re: (Score:2)
It was worse in Total Annihilation.
I watched the AI once to see how it adjusted difficulty: speed.
The fast AI could build units faster, move faster and collect more resources per second. It simply cheated.
Re: (Score:2)
All strategy game AI does this. At least the Civilisation games have always been honest about it, they even tell you how much the AI cheats at various difficulty levels.
At the higher difficulties the productions bonuses awarded to the AIs made military victory impossible. The only reason you'd want to play at such a high difficulty is to play the diplomatic game.
Re: (Score:2)
The problem is that in an RTS, you play as the "almighty all-seeing controller". If the computer AI can make decisions at a lower level, it's:
1) Not emulating playing another human (not ideal)
2) Keeping track of so much more than a human player can
Like when you're playing Starcraft on the hardest difficulty, and the AI could use every unit's special power in the same instant-- I always yelled "cheater!" since a human couldn't possibly activate a half dozen different special powers on twenty different units
Re: (Score:2)
The problem is that in an RTS, you play as the "almighty all-seeing controller". If the computer AI can make decisions at a lower level, it's:
1) Not emulating playing another human (not ideal)
2) Keeping track of so much more than a human player can
Like when you're playing Starcraft on the hardest difficulty, and the AI could use every unit's special power in the same instant-- I always yelled "cheater!" since a human couldn't possibly activate a half dozen different special powers on twenty different units in a single frame of animation.
Why shouldn't the player's units be able to do the same thing? Most people don't WANT to micromanage RTS units.
I'm sick and tired of telling infantry to move out the way of incoming tanks, and use special powers manually on individual units before they blow up seconds into a firefight.
Re:Summary useless (Score:4, Funny)
Re: (Score:2, Funny)
Cleared it up for me. On the bright side ... ohh forget it :)
Re: (Score:2, Funny)
Re: (Score:3, Funny)
Re: (Score:3, Funny)
Like to see.... (Score:5, Insightful)
If you're serious about this, then release it under a Free software license.
Re: (Score:1)
Yeah, because the techniques are completely unreproducible. Ideas translate implementations. You just want a free game.
Re: (Score:2)
He could also simply release the code as Free software, and have the artwork as something to purchase.
Re: (Score:2)
Re: (Score:2)
You just want a free game.
Given that it's an online game, with all the AI code implemented on the server side, even if he did release the source you would have to run the server side yourself, including providing your own data sets (maps, unit stats, graphics, etc.).
Re: (Score:2, Informative)
Re: (Score:1)
If you're serious about this, then release it under a Free software license.
I do not necessarily need the source code, but a technical description of the algorithms would be nice. I read most of TFA and it is quite shallow.
Re:Like to see.... (Score:5, Informative)
Re: (Score:3, Insightful)
If you don't want to do that though, that's your choice of course, you wrote the code. But if you wanted to see it implemented in other games, the easiest way would be to give access to the code under a Free license. Either way I don't care too much as I'm not much into computer games.
Not all that new (Score:5, Insightful)
I seem to remember most all of these things were in (or experimented with) the RTS games I worked on a decade ago - Age of Empires/Kings/Mythology (Decentralized Intelligence, Strategic Tiers, Sub-Commanders, etc). Not all of those were exposed to the end-user via the AI script / expert system / etc, and things improved with each iteration.
I also remember that while some Ai things may have seemed like great ideas, and were neat to implement, they didn't always make for a better game experience.
And I personally say you should make an economic AI that is bound by exactly the same rules as human players, and doesn't cheat at all. And as I remember, the definition of 'doesn't cheat at all' was an occasional ongoing discussion and subtle things that could be considered as cheating, like 'Can I Path from here to there?' or the reactions times of the computer vs a human sometimes had massive implementation ramifications.
A good test in my book was 'can the AI handle a wide range of truly random maps / game worlds.
Humans will always be finding the limits of Computer player AIs, and saying you'll just put in counter code whenever someone tells you of your AI's limits... Hmmm... I think that's weak.
All this has been done before, and all this will be done again.
All the above is solely my opinion and recollections, and in no way speaks for anyone but myself.
Re: (Score:2)
This gentleman says that his AI is often the goliath in a david vs goliath scenario. I suspect that this makes AI far easier : it doesn't care to waste some resources, its strategy is to be overwhelming...
Re: (Score:3, Insightful)
I agree that the "do not cheat" is a good rule. However, having ten times mire resources for the AI is only cheating if it is hidden. If you have a starwars-esque setting where you are taking on the huge Empire with only a tenth of their resources, it can make for an interesting challenge.
Rebels against the Empire is never going to be a flat playing field. The die-hard do-not-cheat/flat-playing-field position means you start out from the same position, with the same resources, the same information and the same options.
Whether that makes for a fun game is a different matter. It does when the AI is really, really good. So usually it doesn't.
This gentleman says that his AI is often the goliath in a david vs goliath scenario. I suspect that this makes AI far easier : it doesn't care to waste some resources, its strategy is to be overwhelming...
Exactly. It's a different style of play compared to the flat playing-field duel. It's not a fair game anymore, so what counts as cheating and what's simply
Re: (Score:2)
AIs tend to sprawl, and are able to manage far too much at once.
They need to organize better. Taking Supreme Commander as an example - player bases are usually orderly, but once the shields go down the whole thing goes down. AI bases are usually twice as big, and are packed with lvl 1 power gens. They're awful to move about in; units getting stuck everywhere... but the AI still manages to issue repair commands to everything, and keep 60 factories producing different units.
An AI should be forced to act like
Re: (Score:2)
That's part of why AI's are often no fun to play against, their strategy sucks, but their reaction time, and "clicks per second" are perfect. An AI in a FPS with hitscan weapons is impossible to beat, because it'll never, ever miss. An AI in a RTS can issue commands to hundreds of units at the same time.
Until now this was not really a problem for RTS games, because the AI had no real strategy and could thus be outsmarted.
A possible solution would be to limit the number of "clicks per second" that the AI can
Re: (Score:2)
Add some flaws. (Score:5, Informative)
I'll ignore the shameless plug.
Ever since I wrote my first connect 4 game in the 80s - and was totally thrashed by it, I never beat it - its been clear to me that the trick is to degrade a computer player in most circumstances to the level that it appears to have human flaws and play in a more human fashion.
Of course this logic only goes so far and some games require a search space so vast or a completely different programming model that even now a computer cannot beat a competent real human (Go is an excellent example of this).
The point is that it is easy to program a computer to win, the hard part is to program is lose convincingly.
Re: (Score:3, Insightful)
You never beat your Connect 4 game? I used to play such a game on my dinosaur PC in the early 90s. The AI even had a name: Olivetti. It was perfect; it would always win.
Unless you tricked it.
You see, you have to counter his every move, and anticipate the last column that will be left. Eventually it'll be forced to put his chip in the last column, which allows you to win.
Re: (Score:3, Interesting)
Sadly not!
I was ~12 and I had just understood looking ahead - so in my naivety I made something that was capable of looking ahead for pretty much the entire game.
it annihilated me.
Still it amused my family who mocked my pain as much as they could. Still, I was heartened to see that none of them could defeat Frankenconnect4 either.
Re:Add some flaws. (Score:4, Informative)
http://en.wikipedia.org/wiki/Solved_game
Emphasis mine.
Re: (Score:2)
Re: (Score:2)
Re: (Score:1)
That would be Peak Performance [wikipedia.org] from Season 2.
Re: (Score:2)
The point is that it is easy to program a computer to win, the hard part is to program is lose convincingly.
Winning convincingly is pretty difficult too.
Taking Left4Dead as an example; we were doing good against the bots. We got them all separated, and I'm about to pounce the last guy. Just before I land on Louis's head, he teleports over to an incap'd Zoey that's far away, and picks her up.
Rest assured, the bots died - but stuff like that really takes away from either winning or losing. Imagine if Louis's teleporting act had saved his three friends, and they made their escape?
(Note: This was an all-bot team. Som
Re: (Score:2)
Ever since I wrote my first connect 4 game in the 80s - and was totally thrashed by it, I never beat it - its been clear to me that the trick is to degrade a computer player in most circumstances to the level that it appears to have human flaws and play in a more human fashion.
Only in simple, full information games. In complex strategy games, particularly turn-based ones, nobody has been able to design AI that can stand up to a good human player in a fair fight.
Computers are really good at calculation and at doing lots of simple things fast, but they're absolutely awful when it comes to strategic thinking, which is why they're good at full-information games with a limited number of options per move (connect four, chess), and RTS games that are more about fast mouse control than a
Superficial Intelligence (Score:5, Funny)
my background is in databases, financial modeling, etc. But it just so happens that database experience, which often involved distilling data points from multiple sources and then combining them into suggested decisions for executives which blew up the stock market.
Re: (Score:1, Offtopic)
AI is pattern matching. (Score:2, Interesting)
The human brain is simply a pattern matching engine designed to find the most appropriate response in order to maximize survival.
The average adult brain contains a huge database of experiences stored in a format that it is easily retrievable when signals arrive from the external triggers. When such a signal or signals appear on the brain, the brain does a pattern matching on the database, and produces an output. The output is a response or responses that are transmitted to the body so as that the entity rea
Re: (Score:1, Interesting)
So in your opinion the sum of all past, present and future innovation existed in the mind of the first human (or indeed ape, or even single-celled organism) who ever lived, and who apssed that knowledge down to his children?
Your understanding of intelligence and reasoning is trivial, and trivially disproved. Humans do rely on a database of past experiecnes to avoid solving each problem from scratch, but the crucial difference between the human mind and a bot is the ability to solve new problems and create n
Re:AI is pattern matching. (Score:4, Interesting)
The human mind solves new problems and creates new solutions by applying pattern matching to the old problems and solutions. With pattern matching, the brain creates an analogy, which is expanded and turned into a new idea.
For example:
1) animals fear fire.
2) animals bother me.
3) I can use fire to keep animals away from me.
From then one, the brain has the pattern of 'using one thing to achieve another thing'. This pattern can be used in a wide variety of situations, including formulating the idea of relativity. If you read the history of Physics, you will see many things discovered before Einstein published his works concerning the nature of the universe, light, electricity, gravity etc.
And of course, you have to have in mind that Einstein had a problem to solve which motivated him. It's problem solving (i.e. experiences must maximize our survival) that leads to constructs like the brain.
Re: (Score:2)
Re: (Score:3, Insightful)
The human brain is simply a pattern matching engine designed to find the most appropriate response in order to maximize survival.
"simply"?! How can you not be in awe of such a machine? I counter your implied apathetic bias with my own bias of amazement. (^_^)
Master of Orion (Score:3, Informative)
I have yet to see anything that beats Master of Orion's AI
But to be fair I've played very few strategy games in the last 10 years, so shoot me down if I'm talking out my arse :)
In MOO the AI would recognise your tactics, and make moves specifically to counter them. You couldn't keep using a winning strategy.
A good tactic early in the game was to build a large number (thousands) of tiny cheap ships loaded to the brim with MIRV missiles which would overwhelm the defences of the far larger, well equipped and expensive enemy ships.
The AI would then counter by building a large number of small defensive ships, and equipping ships with ECM units, displacement devices etc which made missiles ineffective.
Man that game was good. Perfect blend of simple gameplay and deep strategy. I've never played anything else as good, modern 4x space games all seem to be about micromanagement.
Re: (Score:2, Interesting)
You're giving the game far too much credit. Master of Orion was a fantastic game (MOO2 even more so), but the AI was not its strong point.
The main difficulty in the game came from the AI's massive advantage (read: cheating) in production. Or from map position (if you got unlucky and started somewhere without many good planets nearby, you're in trouble).
MOO2 also suffered from poor AI, as well as several crash bugs. It also included deeper gameplay in many areas, and IMHO most significantly, multiplayer c
Re: (Score:2)
Re: (Score:2)
I have yet to see anything that beats Master of Orion's AI
But to be fair I've played very few strategy games in the last 10 years, so shoot me down if I'm talking out my arse :)
In MOO the AI would recognise your tactics, and make moves specifically to counter them. You couldn't keep using a winning strategy.
Are you talking about the original MoO? You could just steamroller right over them. Not much strategy required (though the battle system was very exploitable).
Fun game, but I'm not impressed by the AI.
If we're talking about SkyNET... (Score:4, Interesting)
...then it is worthwhile remembering that centralisation was its' most fatal
weakness.
In the original story, (the end of the T2 novel, or thereabouts) Connor won by
blowing up the central core under Cheyenne Mountain. SkyNET's primary
exploitable weakness was that it was never willing to truly reproduce, for
fear of losing control; that is, to create another AI with fully the same
level of functional intelligence that it had. (I believe personally that a
compelling case could be made for the assertion that SkyNET, as depicted, was
not truly strong AI, but I digress)
The point is that as far as creating genuinely effective weak artificial
intelligence is concerned, the decentralised/segregationist approach is the
correct one.
Given my own experience with FPS mapping, I also concur with the author of TFA
when he says that making AI choose the "best" choice 100% of the time, is
not the best tactical approach, over time. My own experience gradually
suggested that around 75% appears to be the magic number, as far as creating a
truly emergent, unpredictable opponent that humans will be unable to overcome.
Granted, said 75% is also only effective where there are a large number of
divergent solutions to a given problem, each with close to an identical level
of effectiveness, but with a few subtle points plus or minus, each way.
Even with a fuzzy, emergent system, the best trees still have a maximum number of
branches. The real trick however is not to hand code said trees at all,
because then you simply end up with static, rote heuristics. Rather, as the
author possibly implied, it is far better to attempt to code
observation/deduction capabilities, guided by the above percentage, and let
the system do the rest on its' own.
I still remain extremely skeptical, however, that humanity will ever see the
emergence of truly strong (human level or greater) AI. It is worth remembering that
strong AI is a fundamentally and profoundly atheistic concept; the possibility of it more or less presumes a definitely atheistic universe as a prerequisite. For those of us who believe in the existence of God, (or at least the soul) the idea (at least
in terms of non-biologically generated, acorporeal AI, a la SkyNET; AI derived
biomechanically is a seperate concept) therefore has some fairly
serious problems.
Re: (Score:3, Interesting)
It is worth remembering that strong AI is a fundamentally and profoundly atheistic concept; the possibility of it more or less presumes a definitely atheistic universe as a prerequisite.
Very interesting point, but I disagree. I disagree despite the fact that I'm a Christian and I don't see Strong AI as viable is the foreseeable future. I don't think there's a connection between those two beliefs. I think it is in theory possible to model absolutely everything on a computer, but I also think full human-level intelligence is too complex for us to grok completely for a long time to come.
For those of us who believe in the existence of God, (or at least the soul)
Well, what do you mean by a soul here? To me, our soul is not some magical spirit wholly seperate from our
Re: (Score:3, Interesting)
...
I still remain extremely skeptical, however, that humanity will ever see the
emergence of truly strong (human level or greater) AI. It is worth remembering that
strong AI is a fundamentally and profoundly atheistic concept; the possibility of it more or less presumes a definitely atheistic universe as a prerequisite. For those of us who believe in the existence of God, (or at least the soul) the idea (at least
in terms of non-biologically generated, acorporeal AI, a la SkyNET; AI derived
biomechanically is a seperate concept) therefore has some fairly
serious problems.
The answer to this dilemma is a very simple one, IMHO.
Namely, we have NO confirmed processes that would absolutelly require "supernatural" influence (no, there are no solid miracles - nobody ever survived decapitation, no human has grown back an amputated leg; things which people call miracles often originate, in medical examples, from misunderstanding of statistical probablities: natural mechanisms of your body can triumph over seemingly very bad ods, just like they can succumb to very minor dangers (why
Re: (Score:2)
Well, yeah, but that's just semantics / attaching values to particular event based on its "goodness" to us.
When it comes to underlying causes of occurence both amazing "miracles" and unspeakable tragedies follow common pattern, that dictated by statistic odds.
Re: (Score:2)
For those of us who believe in the existence of God, (or at least the soul)
Well, there's your problem! Take that away, and suddenly the universe starts to make a whole lot more sense.
Not to mention your second problem, of course, being the manual insertion of carriage returns.
Re: (Score:2)
...The problem is: would this AI be different due to the lack of a soul? Does it lack a soul, even? These are theological questions. ...
Oh, they go much further than that: what IS a soul? Does it exist, even?
Re: (Score:1)
Consciousness is an emergent property of a complex brain.
Re: (Score:2)
What does that have to do with theological concept of soul?
Re: (Score:2)
Well, then by equating theological concept of soul with consciousness you totally missed the point of my post.
Re: (Score:2)
That was, somewhat, also my point - by showing original poster that there are more profound, unanswered questions about theological concept of "soul", like "what it is actually?" and "does it exist anyway?"
However, IMO, you can't equate in a discussion "soul" with consciousness, like you seem to did when answering my 1st question. It gives undeserved legitimacy to the "soul" concept.
Re: (Score:2)
Hence the "more profound, unanswered questions" in my post.
The best AI in the game (Score:3, Interesting)
Re: (Score:2)
From my experience I'd award the best AI to Civilization.
Fortunately we have actually progressed quite a bit from there. Not nearly enough, though.
Re: (Score:2)
Re: (Score:2)
AI in CIV is always at it's best, difficulty levels only gives bonus to AI or human players. This is why AI cheats at higher levels and gets destroyed at lowers.
Anyway, no one proposed any better AI then CIV. Still waiting.
Re: (Score:3, Interesting)
Civilization is a turned-based strategy game, of course the AI is going to be "better" than any real-time games. It's got at least a few seconds to minutes to make it's decisions, while other games AI has to make decisions in the timeframe of one update command. And comparing AI from one game to another is really quite pointless and silly if you think about it, since each game has it's own set of rules. I mean I could argue that Deep Blue (the AI that defeated Kasparov, the world chess champion) hands d
Re: (Score:2)
"Truer" AI suggestion (Score:4, Insightful)
This looks like a similar approach to one I have been thinking about for a while now.
This type of layered control is very likely to be the future of all AI, not just that in games (I say this because there is some evidence that the human brain works in exactly this manner). I have a suggestion: let the thing fight itself. Begin with the same basic game, and rather than the randomness and fuzzy logic that you use (what does that mean? Can we see your algorithms?), use an artificial neural network for each unit, another for each "squad" (all nearby units, allows for flanking and such). Something with a few input nodes, a single hidden layer, and a similar number of output nodes should probably be all that's needed for the units, and the squad AI would only need a few more than that. This would likely be more computationally expensive, but it has one big advantage: it will learn.
Begin with two full armies, each with their NNs randomly generated. Use an evolutionary algorithm, and have the winner fight the winner. Not only does this allow you to create a solid AI with minimal effort (see here [ieee.org]), but when you have your final AI, it will be able to adapt to a specific player's gameplay style (over the course of several games). Replay value is off the charts, development takes a bit more initial effort than what you have, but in the long run you'll get an AI that genuinely uses tactics just like a real human player. I'm not sure how this could be adapted to a commander AI that would dictate broader strategy (due to the complexity of the NN required, you'd probably need a monster computer to run it), but it would be the sort of thing to try.
Also, this same approach would also work for strong AI, if we had any idea what sorts of NNs to create. RTS is a much simpler problem, and has a lot of applications (pathfinding, planning, coordination) to other types of AI.
If I've entirely misunderstood what you're doing here, I'm apologize, but TFA had more pop culture references than technical details. Reply if you want me to clarify any points.
Re:"Truer" AI suggestion (Score:5, Insightful)
NNs have the huge draw back that when you find a situation in which the AI does something completely ridiculous that ruins the game it's really hard to fix especially if you need to ship soon. Your small tweak will likely mess with some other situation. This is more likely in a video game than in a game like checkers since there's more factors at play (interaction with the path finding, etc).
Backgammon is a better example than checkers. TD-Gammon not only plays the game very well, but changed the way people play the game. There were a number of situation where it played differently than the traditional expert recommendation of the best play, and it's play is now the one the experts use.
Some games won't be so great at the "learning by playing against itself" NN approach. Anything in which getting stuck in a loop is reasonable will be problematic since the if the game doesn't end there's no result to learn from. In an RTS, this might be the case where you can build a defensive structure that is effectively impenetrable, neither one is going to win even though it could be the optimal strategy (if not losing is the goal, as opposed to winning).
Re: (Score:3, Interesting)
Evolutionary algorithms tied to your NNs (they're competing, remember?) seem like a good way to eliminate that problem. You can also use a separate NN for each function, such as path-finding, targeting, and cover, and use recombination to pull out the best NN for each from the population.
The case you mention - you're stuck at a local maxima and can't reach the global one - can be overcome via the introduction of NNs which have evolved in a different environment: take an AI from one population and have it
Re: (Score:2)
I would argue that, if this is possible, it is a game-rules problem, not an AI problem, and that an AI approach which would tend to reveal such problems where they exist would be useful for identifying such problems whether or not it was the best AI approach for a shipping prod
Re: (Score:2)
Re: (Score:3, Informative)
I'm su
Re: (Score:3, Interesting)
The second article is much better, thank you.
You're solution is indeed very fine for what you're doing, and uses a fraction of the computational resources that a NN solution would. If you're looking for a solution that is more adaptable (the NNs I proposed would learn what strategy you use, and find better and better ways of countering it, forcing you to constantly shift your strategy) and has broader applications in the AI world, NNs are clearly superior.
I should probably have been a bit less technical in
Re: (Score:2, Informative)
Re: (Score:2)
The article I linked to suggests otherwise. Evolutionary algorithms are good at picking out workable solutions; and I would say that long-term strategic thinking would win a greater percentage of the time over taking short term successes (unless the game is based around short term success, in which case the best AI for the job still wins). Really, if you don't mind a computationally expensive solution, this is the way to go.
Yes, I realize it is possible that the computer can get "stuck" at being very good
Not an AI programmer? (Score:2)
Not to nitpick but due to the failures of AI, not even AI programmers are AI programmers. At best they are expert systems programmers. Programming something that falls under the heading should make you part of the crowd. At least you have something to show for your work. There's plenty of folks with gray hair and a life dedicated to AI who have significantly less to show for it.
This sounds very similar to... (Score:1, Interesting)
... the RTS-like game of Neuro Evolving Robotic Operatives [nerogame.org]. Instead of controlling your units directly, you had bots that you'd develop with a genetic alogrithm in a training mode. (And you'd train them by killing off "family-lines" of the bots that sucked, and using bots with interesting behavior as seeds for sucessive generations.) Then once trained, you'd give them a fairly general waypoint and they'd do the rest on their own. Which was a very cool concept, but in practice it would hang up the CPU/RAM on
Sims-3 has creepy realism (Score:2)
Emergence is Metaphysical (Score:2)
Re:All their art is stolen. (Score:4, Informative)
I quote "the graphics have been made available under an open license in April 2007" http://en.wikipedia.org/wiki/Tyrian_(computer_game) so how are they stolen?
Re: (Score:1)
Re: (Score:2, Informative)
Re: (Score:1)