AI behaviors

[marhawkman]

Since AI will be getting implemented eventually, we need to decide what behaviors it will have.

So far I've thought of a few things.
1: have them dislike you for attacking them. but not necessarily immediately declare war.

2: ditto for going around in their space, but only if you aren't friends with them. It's so f'ing annoying how simply cohabiting with ai's in MoO3 and SE3 causes them to hate you and decide to get rid of you.

And of course a few other things, like some being more aggressive than others.

[ewh02b]

pehaps have a -5 to +5 system, where your relations with everyone start out at 0, and as you interact with them, the relationship gets better or worse?
kinda like the sliding scale in Moo2, but easier to understand.

-5: total war. they hate you with their heart of hearts, and focus all their resources on destroying you. only one player can have a -5 from an AI at a time--they can't focus all their resources on multiple players.
-4: at war. they hate you, and will attack your fleets and your planets.
-3: half-war (piracy and such). they do not tolerate you, and will attack any fleets of yours, as well as any planets sharing a system with them.
-2: unfriendly. they tolerate you, and leaving a fleet in a system that they have a planet in will soon degrade the relationship to -3. they will not attack your planets.
-1: untrusting. they tolerate you, but refuse any sort of alliance with you. will allow your fleets passage, but will protest (diplomatically). will accept bribes and other forms of diplomacy to raise their rating of you.
0: neutral. perhaps open to a trade alliance, if a bribe is offered in exchange.
+1: will accept a trade alliance, and will allow your fleets passage.
+2: will accept a research alliance.
+3: will accept a mutual protection agreement.
+4: mutual protection. will accept an alliance, and will occasionally pass intelligence on to you (spying).
+5: alliance. will attack your enemies and share intelligence freely with you.
===
I'd say that an attack would be immediate grounds for a declaration of war--wouldn't you declare on someone who attacked you? If you attacked them, it would probably drop to -3 immediately. another attack, -4. continued warfare, perhaps some planetary bomboardment or other atrocities, and drops to -5. At -5, there's no where to go--one of you has to die. -5 is Hitler vs. Churchill.

at -4, you probably have to offer a bribe (planetary system or tribute) to get them to -1--I don't think you could raise it higher in a single turn. at -3/-2, you would probably have to give them a one-time payment, perhaps some research, to go to -1. at -1, some time is needed to raise it to 0, or a gift can speed it up.

[utilae]

You should be able to gain some love with all comp players for acceptin, starting a treaty. You should gain some hate for breaking a treaty. So if you are constantly breaking treaty's (eg kill a few planets, enemy offers treaty, you have break, then break treaty kill more planets), it will be hardy to have a treaty with another race because it doesn't trust that you will keep to the treaty for long.

[Krikkitone]

Well an important Decision that need to be made with the AI is whether thay are

Characters

OR

Players


A Character will attack you because its the type of thing they do at this point in the game. A Player will attack you because all indications are that's the best way for them to win the game (or get a good score) at this point.

Now admittedly the AI won't be able to be as good of a Player as a Human will, but if they are 'Designed' as Players then they will be better.

For example a Player, if designed to win, would NEVER vote for someone else in a Diplomatic Victory (they might if they didn't believe the someone else would win and they hoped to get some favor by it.) A Character might.

A Character might stick with an Ally because they have been good to them for the entire game. A Player would Backstab that Ally if they thought it was good for them... regardless of how long they had been allies.

Part of this can be avoided by making 'Winning' a more flexible idea.... so that you can 'Win' as a group.

Other ways include the happiness of 'populations' with foreign policy and foreign powers. (so that your people might be unhappy with a trade treaty with someone who attacked you a bit earlier) But essentially those are designing the Game so that certain 'styles' of play become good for players.

The AI should first and foremost be a Player, (until it comes time to adjust for difficulty levels, then 'Character' begins to overwhelm it.)

[MikkoM]

I think that the two most important things in AI behavior are

A. The AI should act like the race it represents.

So if the AI is a peace loving race then it shouldn`t be attacking your fleets and planets at the every possible moment.

B. The AI should act like the military and political situation of the galaxy requires it to act.

This would mean that even a peace loving race could become quite aggressive if it needs, lets say more systems to protect it self from a growing hostile alliance.


Maybe these two things are quite obvious, but I have seen too many games were the AI for example when offered to surrender to you when you have clearly defeated it refuses to do that and just keeps fighting the hopeless war against you. Now of course there could be some species, which would rather die fighting than surrender, but I don`t think all species would be willing to face extinction rather than surrender to you.

And what comes to A. I think that war loving species could be little more attracted to each other than to the peace loving species and the peace loving species would then naturally be little more attracted to each other than to the war loving species. This would mean that it would be little easier if you were a peace loving species to make treaties with other peace loving species than with the war loving ones, because the war loving species might not think so much of your peaceful talks and behavior.

Then when the AI starts to colonize worlds and explore the galaxy the scout engagements and the contests over systems could effect the relation ships of the empires, as if your scout is destroyed you wouldn`t probably be too happy about it, but still maybe some of the war loving species might want to test the abilities of your ships and so attack them. Also the contest over the systems could have a negative effect on relationships as you wouldn`t probably like to see some other empires colony ship colonizing the planet just before you do.

Also using one empires space to move your fleets could have a negative effect on the relationship between the empire whose space has been used and the empire which used it, unless they are allied to each other or have made a treaty which gives the fleet moving empire the right to use the space of the other empire.

Then when you make treaties with some empires, this should have some sort of effect with other empires that have made threaties with you or with the empire that you have made the deal with. This could for example be seen as a negative reaction with a species that has become quite friendly with you, but has cold relationships with the empire that you made the treaty with. So their liking towards you would drop a little.

Naturally the alliances and wars would have the biggest effect on AI relationships. Now if an empire would declare war on you the other enemies of the empire that has declared war on you might become friendlier to you as you would have a common enemy and so you could form strange alliences with species that have earlier been very cold towards you.

And of course if you make an alliance with some one it sends a clear message to other empires that you are now very friendly with that empire which you have an alliance with. So some of your former friends might not accept the other empire and so become more colder towards you.

Also there could be some preemptive threaties, for example between to neighbours of a species that has become very powerful and aggressive in the galaxy.

Now of course the aggressivneness, the point when the war is declared, the willingness to peace etc. would depend on what kind of species the AI represents.

Maybe alot of these things are about diplomacy, but as I think diplomacy is one of the most important parts of AI behavior I think they should be brought up in this conversation. I also used a lot of you forms, because it is easier to demonstrate my ideas in this form.

[Krikkitone]

A. The AI should act like the race it represents.

Well here I think the danger comes in.

The BEST way to do ths is

1. Make sure the game allows several different equally Successful Strategies/Approaches to Win.

2. Make a race's picks be one of those things that helps tilt the balance towartds or away from various strategies.

3. Make an AI pick particular strategies based on the situation, Including the picks of their race.


So a race that had bonuses to tech research would tend to make its AI go for Tech friendly Strategies, Race's with Tolerant/Survival/Food bonuses would emphasize expansion over development, etc.

Essentialy all AIs would be the same. However, the AIs would develop 'long term strategies' like going for Conquest this game, going for Development, etc. They would then tend to stick to that strategy until the situation changed enough for another strategy to look Much Better.

[Impaler]

I have been thinking a bit about diplomacy and relations and have concluded that the traditonal Love-Hate scale of diplomacy is fundamentaly flawed.

It cant take into account the many nuances of true diplomacy and is terrible at recognizing differences is size/power between empires.

I propose a MULI-SCALAR relationship, in addition it would be BI-DIRECTIONAL. Let me elaborate, Empire A will have several scales that express how they feel about Empires B. Empire B would have another completly DIFFERENT set of scales to reflect their feelings about empire A.

I am thinking 4 key scales that reflect the most important things in a relationship and what states do to each other. They are...

POWER - This is a military strength assesment of your empire vs their empire, it includes assesment of your economic strength not simply your current ships on hand. If they percive you too be more powerfull then them then they will desire to alie with your or opose you by creation of defensive aliances (if hopelessly overpowered they might just surrender). An Empire which is stronger desires to vasselize, concour or allie with weaker powers. If either side has alies and pacts then these can come in and alter the assesment, a modest power that is part of a strong pact sees itself as more powerfull. The same applies the other way as well and in both cases the solidarity of the pact is very important, alies that cant be relied on are judged as useless.


THREAT - Threat is related too be destinct from power. Ware power is simply the ability of one empire to smack down another threat is their percived liklyhood that such a thing will actualy happen. Physical proximity, verbal and physical threats (aka fleet buildup on borders) and a reputation as a war monger increese Threat. Being ocupied in a draining war effort withsomeone else, non-aggresion pacts and other goodies (trade, research, inteligence pacts) that would be lost if war were to breakout decresse Threat. Threat is sligtly modified by low power (suicidal wars of agression against vastly supperior empires is uncommon) and good mood (see next section). All empires act to try to reduce the threat other empires pose too them (this could be done in peacefull or none peacefull ways). Power x Threat gives FEAR which is one of the major determinants in going to war.


MOOD - This reflects all hostile/friendy actions actualy taken against an empire. Think of this as the mood of the street, do the people have some reason to realy hate you or love you. Ware fear reflects an empires thoughts about the future, this scale is conserned only with the past, some races keep grudges, others forget quickly. When you perform actions that are good/bad towards a race the amount of movment here is modified by how much they were expecting that action (suprises always incresse the effect of the action), and they take into account how much that action 'costed' you. Say you side with them in a war against a powerfull foe they respect the sticking your neck out factor of your actions.


TRUST - This is basicaly a 'do you keep your word' assesment. If you break your word it goes down. If you keep your word it goes up. Many games like SMAC start you off with 'perfect' reputation which can only degrade over time. I think a more nuanced aproatch is called for, you start off nutral (well perhaps some modifiers apply say for Alkari vs Mrsharans) and you can incresse or decreese you level from that point. Also note that when you break your word you suffer loss of trust with everyone and loss double with the ones you betrayed. Some races are very honorable and almost never break their word, others are total backstabers, some races are trusting and easily buttered up, others are very slow to give their trust. As with Love/hate they take into account how 'costly' it was to keep your word. Trust is important for all players not just peacefull ones, if you are belived your threats are more effective (if you threaten and are rebuffed and do not retaliate you loss trust) and you can negotiate surrenders more effectivly (otherwise they fight to the death)

In addition to a current position on each meter their is also a 'Change' value for each meter that reflects what direction its moving in and how fast (aka 'falling rapidly', 'stagnant' ect). This will alow empires to notice things like rapid power incresse BEFORE they are overtaken. Basicaly change in a direction up or down accentuates the meter in that direction giving the AI some limited forsight. As people have speculated we should simply have 1 AI which makes desisions based on the present situation and plays to its strengths at the time what ever they might be. The warmonger race will go to war not because its script blindly tells it to do so but because the script sees that they have a comparative advantage and war is the 'smart' thing to do. To give a race personality were would have modifiers much like Moo1 had, they effect the readout values of the 4 meters I propose. For example a 'cocky' Empire gets a bost to its own power assesment and this slightly inflated opinion of itself is feed into the AI script which then makes desisions based on that number. The AI behavior will be pushed slightly away from a total unbiased assesment but it will still acknolage the reality of the game environment and react in a coherant way.

[MikkoM]

A. The AI should act like the race it represents.

Well here I think the danger comes in.

Yeah, I also thought that just having the part A. could be quite dangerous and that`s why there is also part B. (The AI should act like the military and political situation of the galaxy requires it to act.) in my original message.

My idea behind that text is that the AI empires would enter the game behaving, in the early stages of the game, like they are designed to behave. So if an empire is technology minded then at the early stages of the game it would use a lot of its resources to the technological development.

This would be realistic as I don`t think for example a war loving species which has a long history of its own violent traditions on the surface of its own home planet would just suddenly choose to become an agricultural, peace loving race because it entered space.

Now what would cause the changes in the races strategy would be its contacts with other races as the political and military situation of the galaxy were the race lives naturally affects its decision making. So like I said before, this could make for example a peace loving species to become quite aggressive, because it needs more space to get resources and build its fleets against a hostile alliance, or a war loving species would suddenly focus on technological development because it has noticed that it is falling behind in that area.

But what I think would still be very important would be that races had their "souls" all game long. This would mean that a war loving race wouldn`t transform permanently to a peace loving, flower picking one.

Also if the species aren`t going to have some basic characteristics which might of course be hidden for some time or they might be soften a little, so that a war loving and its own superiority thinking species might at the end of the game be a little more peaceful and tolerant to other species as well, then what`s the point of making these alien species at all if they really just are the same one species masked underneath different kinds of appearances?

[marhawkman]

Well the differences could just be how much one of the scales needs to be tipped in a cetain direction for it to affect them.

On a side note.... We need to make it so that havin a small negative relation with an empire for several years doesn't cause them to eventually declare holy war.MoO3 and SE3 are really bad about this...

[cubic1271]

Howdy ya'll. Hope you don't mind the length of this post, but was sitting here brainstorming about how an empire-wide AI might work and thought this would be the thread to post it in. I do apologize for the detail, but I felt that examples help to clarify the concept. That being said, here goes. . .

***WARNING: VERY LONG POST***

I haven't really had the chance to take a look at the game's source yet (waiting for it to download while I'm writing this post), so some of the methods I suggest might not be feasible / whatever. If so, can just add code to make them feasible (ain't open source grand? ).

That being said, here's a method that I think might work for an empire wide AI, broken down into steps that the AI would follow and how the AI would evaluate its position on the map and what have you. I'll be breaking down the AI's turn into three distinct phases: study, plan, execute

I - Study

This is the phase when the AI looks at what its opponents have done and tries to interpret it. Because it's really hard to examine and interpret every move that a human makes, perhaps the AI could look at:

1) Distribution of firepower

This category would be a weighted value, corresponding to the amount of firepower a player could bring to bear on the closest AI-controlled system, tempered by the distance each ship is from that system. A high number on a large number of the AI's sectors means that an attack from that direction is likely.

For example, say that a player controlled 30 Mk. I. assault ships and had 18 of them in a sector adjacent to an AI's system, and another 4 in sectors with distance=2 from that same AI-controlled system, and the rest had distance=4 from that AI-controlled system, the value would be the sum of all ship's firepower rating / their respective distances from a single system. In this case, let's say the firepower of a single assault ship is 5. We would have (5*18)/1 + (5*4)/2 + (5*8)/4 = 110 against that sector.

Once we have found this value, we would run the same algorithm again, except this time we would run it for the AI's ships and subtract that value from the value we found for the sector (we'd also add one to the distance to reflect the fact that ships need to be in a sector to defend it). Say for the sake of argument that the AI has 31 ships, 15 of which with a distance=1 (they're in the sector), 4 ships with a distance of 2, 7 ships with a distance of 3, and the rest with a distance of 4, we'd have (5*15)/1 + (5*4)/2 + (5*7)/3 + (5*5)/4 = 75 + 10 + 11.6 + 6.25 ~= 103

These number could be saved for a few turns to study shifts in fleet distribution (indicate attacks, etc.)

The net firepower for that system would be -7, or the AI would not be able to get as much firepower to that system as quickly as a human could. By studying the distribution changes as mentioned above, the AI could watch for fleet shifts that might signal attacks and respond appropriately.

Perhaps once a fleet of ships have commited to a space lane, they could be treated as if they were already in their destination sector for the purposes of this method?

Time to run is number of ships * number of border systems (systems not surrounded by sectors controlled by the same player) * time it takes to find shortest distance from a ship to any given sector

2) Cultural analysis (if culture will play a role in the final game)

Culture would be better handled by individual systems in my opinion, but perhaps the AI could handle it in the following way:

Record the culture gains by player sectors and record how they've changed (average gained over the past 10 turns, record that for the past X turns to watch for cultural attacks on a given sector. To determine attacks, could normalize with respect to what has been gained over the past few turns (I.E. if sector A has seen an increase of 500 culture points by its neighboring sector, but sector B has only seen an increase of 20, it's probable that there's something weird going on against sector A).

3) Sector weight

Watching these distributions is all well and good, but the AI will need to make decisions when it all comes down to it. What should it fight for? What shouldn't it? That's where sector weights come into play.

I would propose that a sector be weighted by the % of all ships that have come from that sector both over the past X turns along with game-long along with the % of empire-wide production that comes from that sector.

Also, border sectors would be of higher weight than non-border sectors; it's easier to keep a solid perimeter than it is to have to go hunting through every system you own while looking for people that may have slipped past your defenses. Or at least that's how it should work in theory

4) Taking all this crap and doing something with it

By analyzing all this information, the AI can determine:

I) How aggressive / passive every player seems to be in various ways (high distribution of firepower over a small number of planets could imply that somebody probably likes to expand slowly)
II) What a player's overall strategy is (opponent is aggressive expand, passive culture. . . play defensively and try to win the person's people over with the intergalactic statue of liberty or perhaps the Pyramids at Polaris?)
III) What the state of the galaxy is. If force distributions are both moving toward each others' borders, for example, you could probably assume that two empires are building up for war.

II - Plan

The AI uses this phase to consult its personality and look up a few things:
A) "Should I be proactive or reactive (aggressive / passive)?"
B) "How certain am I that a player is doing X?" (learning component of the AI, perhaps? but I'll save that for another post)
C) "Okay, if player X is doing this, what does my schema tell me to do?"

So, using the information it gathered from step 4) above, the AI begins to think (or something like that). For example, if the AI notices that forces are moving toward each other and there's a relatively undefended border nearby, what's to stop it from declaring war on somebody and helping itself to a few planets before negotiating for peace again? ( that particular strategy works great vs. AI players in civ; might be fun to turn the tables ) It creates high-level plans with relation to each player, and then determines the best way to carry them out.

For example, say that player A and player B are at war. player C, our AI player for the moment, sees that player A left an open border to its galactic north. Next turn, BAM! player C invades the open northern border.

More specifically, player C notes that the aforementioned firepower distribution favors it by some margin and then goes. These margins could be largely determined by a player's personality (I.E. an aggressive species would be more likely to pull off a lightning-fast invasion and blow through five systems, and might be able to twist an opponent's arm to get a cease-fire again). Also, an AI might simply contact a player and say, "Hey, you idiot, you left your borders open. Pay me X / Give me X and I'll let you live."

Note that these plans are long-term plans, which is why this plan is seperated from the Execute phase below.

III - Execute

Execute is when an AI player can take the plans it formulates and make 'em happen. Much of this is relegated to independent governor AIs (like the kind in alpha centauri or whatnot) simply because those particular AIs should be tuned such that micromanagement isn't required. Ideally, anyway.

This phase will example what was determined during the Plan phase and determine exactly how to do it / how much should be done. Here's where tree searches come in handy; by analyzing a few moves deep, we can come up with an idea of how many ships we can produce before a war ends, etc. Prediction will play a big role in this process. To make it a little easier on the AI players, perhaps the whole AI-runs-on-the-player's-turn concept could be borrowed from GalCiv. Otherwise, tree searches could take a while (depth 4 search with so many possible moves is a scary thought). You could hack off parts of the search tree with heuristics and probabilities, but there's still a lot to look at there. Don't get me wrong, though; trees are the best way I know of to get a semi-accurate picture of what's going in the short term. Long term relations don't change quite so quickly, so it's a whole lot easier to get away with not using trees in the Plan phase.

There is one more vital component to throw into the mix: trust.

In my mind, trust would factor into the Study phase. . . treaties would increase trust and give bonuses to the various computed statistics perhaps, while previous broken treaties would take away from that up to a certain point. Trust would be easier to regain at easier difficulty levels, and some personalities would naturally be more trusting than others.

This could be implemented as a trust table contained by an AI player that contains the current trust level for any other player. Higher trust would be worse, but any given trust entry for another player would decrease at a rate corresponding to the difficulty level of the AI along with their personality.

Diplomacy would be largely related to trust. A larger trust value would mean that somebody isn't as trustworthy, and therefore shouldn't be trusted. Also, perhaps to encourage cooperation between species, intergalactic goodwill, etc., you could introduce the concept of a contract.

Contracts would be enforced by the Senate (the body mentioned in another post somewhere). If a contract were to be broken, you could introduce penalties that would be stipulated at the time the contract was signed (I.E. a peace treaty between two nations, with the stipulation that, should either side break it, they would be responsible for reparations paid to the Senate AND to the victimized party in the amount of X. . . a way to get around the whole 'never trust someone again' thing). People would the option to A) withdraw from the senate, a decision causing them to be slapped with a trade embargo + exile from the senate for a set number of turns or B) pay the reparations in turn-ly increments.

Just ideas / concepts, some of which were borrowed from games (think I mentioned it when I did that, though I can't be certain; sorry if I missed something!) Hope this is useful to somebody.

Also, great job so far! Game looks neat.

[Airshipjones]

I don't think the AI, just like the other players, should have special knowledge of the players assets and movements. I think that knowledge of technologies, ships, their capabilities, locations, movements, or just about anything else. I think that sensors should allow one to estimate ship mass, power output, speed, weapon strength (when fired), shield strength, and maybe a few other things. I think that any other info should be from spies or allies and be no more reliable or timely than what we have on the other players.

[utilae]

Ok, I have a few thougts.

A) Who would send the news?
*It depends which player sets up their own interstellar news organisation. Or even if the players citizens media can broadcast news that far to be pciked up by aliens.
*Why you would want to send it, I don't know other than citizen morale.

B) Where would that news come from?
*Now if news comes from another race, it is most likely propaganda.
If news comes from your spy agencies, then you at least know that's the truth.

So, which news should be displayed to the player. Probably the news from their spy's, though I guess it would be interesting to recieve propaganda and compare with spy news, then realise, hey you "lied" Race A, I declare War.

[Airshipjones]

It would be nice to have spies reporting info on disinformation campaigns by other races. Knowing what they are lying about tells what is important to them, since they are lying about it. If they lie about productivity, or farming output, then they may have a weakness there, or they may be growing faster than they want to admit.

[eleazar]

Well an important Decision that need to be made with the AI is whether thay are

Characters

OR

Players


A Character will attack you because its the type of thing they do at this point in the game. A Player will attack you because all indications are that's the best way for them to win the game (or get a good score) at this point....

For example a Player, if designed to win, would NEVER vote for someone else in a Diplomatic Victory (they might if they didn't believe the someone else would win and they hoped to get some favor by it.) A Character might.

A Character might stick with an Ally because they have been good to them for the entire game. A Player would Backstab that Ally if they thought it was good for them... regardless of how long they had been allies.

Part of this can be avoided by making 'Winning' a more flexible idea.... so that you can 'Win' as a group.

Other ways include the happiness of 'populations' with foreign policy and foreign powers. (so that your people might be unhappy with a trade treaty with someone who attacked you a bit earlier) But essentially those are designing the Game so that certain 'styles' of play become good for players.

The AI should first and foremost be a Player, (until it comes time to adjust for difficulty levels, then 'Character' begins to overwhelm it.)

Yeah, this is an old thread, but it contains an important question.
I agree that the emperors (AI and human) should be "Players", but i propose that the citizens of all empires should be "Characters".

Thus the best of both worlds: the AI Emperor is not unfairly shackled with a "personality" which may keep his empire from winning, but each empire has a distinct flavor based on the nature of it's citizens.
So the Zukbag Devourers would generally be a xenophobic, warlike empire if played successfully by human or AI.

[Tortanick]

So what you're saying is that the citizens of that empire follow their personality, a xenophobic race will see unrest if you're not at war, a flower picking race will create unrest if you are at war.

However the emperors are enlightened players who understand a bit more than the man on their streets, they make decisions based on something like Impaler's scales, however they also count the effect of their decisions on unrest.


I really really really like it! It would need very careful balanceing to hit that sweet spot between behaving like they should and being an incompetent AI leader but if its done right it its an amazingly good system. Especially since it binds players to playing like their race should be played.