-
@redrum wow Good Action Here !
Was really hoping #3 was gonna happen, but at least I'll sleep better : )
-
Some thoughts about "strafing".
I don't know what's already in the AI and what you already have planned, but not implemented yet. Here is some feedback based on my observations of Hard AI in the latest stabile TripleA version (1.9.0.0.13066).
I'd like to think for AIplayer1 to make a strafing attack (negative TUV attack) to set up a follow-up attack from AIplayer2, there are a few things that the AI could improve on to evaluate if it's a good idea or not. The easiest to implement, I figure would be simulating "total TUV" for all attacks. So, the odds calculator for the AI needs to run the first attack TUV and immediately the 2nd attack TUV (starting with units remaining after 1st attack) and add the TUVs together and then repeat for however many simulations runs the AI uses for it to be statistically significant. The average TUV swing can then be compared with what you'd get if ONLY AIplayer2 attacks. If the total TUV swing is not improved by AIplayer1's potential strafing attack, then the AI should consider not making that strafing attack.
So essentially what I'm saying is there should be an odds calculator that can consider several attacks in succession (or at the very least 2 attacks). I imagine if a bit more generalized this could also be used to consider potential counter-attacks by enemies or even combinations of strafing attacks, friendly follow-up attacks and enemy counter-attacks, but obviously it gets more complicated the more attacks are involved as you for every attack need to predict if that player chooses to either not attack at all or commit some troops elsewhere, or if their troops will even be there come their turn, which might not be easy to determine by just looking at battles in 1 territory.
-
@Quorthon4 I do not know the technicalities of what you posted but I do want to re-mention something I've said a bunch of times about a "learning AI" which for me was simply adding a saved "WinnersCapturedTerritories" file to be used by the AI every game for direction of focus which would ofcourse improve the AI as more and more games are played. This could even be "machine learning" which as I described is not too complicated and not require that much more resources other than a saved file and some memory space ... so as pertaining to your post ... I think giving the AI direction and goals in its movements would improve the AI more than just letting it decide by TUV what move to make next!
This thread is about improving the AI and I've yet to see anyone give a better suggestion than mine but let's wait for the devs response to see if your idea or mine are in the future for this AI!
-
@Captain-Crunch first i heard about that idea but i was a little late to the new forum party, anyway, i like it!
-
@Quorthon4 Good thoughts. The AI does some limiting strafing already and essentially follows a process similar to what you describe. It primarily just looks to use it on high value territories like capitals and factories. If you test revised or v3 and give USA and UK large forces next to Germany's capital where neither is large enough to capture by themselves but if they both attack then they can, you should see the AI strafe to capture it.
There are a few challenges with expanding its usage:
- Its slow as you are running multiple BC runs for a single battle so using it across all potential battles could slow the AI considerably especially on large maps.
- Adding when to retreat into it so that the AI would say just use Player1 to attack for a 1 round to weaken it then have Player2 attack to capture it instead of Player1 attacking til all their units are dead. This quickly multiplies the number of potential scenarios to calculate and complicates which is ideal.
- Strafing has different uses in dice vs LL. In LL, you can guarantee a certain number of hits and always know you can retreat where this isn't the case for dice. This often becomes the case when you have a large unit stack then your opponent but are defending your capital so need to leave most of your units there so you just attack for a round or 2 to inflict casualties then retreat back.
If you have any save games showing scenarios that the AI could greatly benefit from strafing please upload them here so I can take a look.
TripleA Developer with a Passion for AI: https://forums.triplea-game.org/topic/105/ai-development-discussion-and-feedback
-
@redrum @Quorthon4 i've seen AI strafe, does it well in 1914 domination map in fact, particularly in the eastern front where Germany and Austria basically need to team up to overpower Russia unless either ignores their western fronts.
-
@redrum I have a saved game where AI uses strafing in a situation where it's highly questionable. Especially look at Austria turn 5 and Ottomans turn 5 (both attacking Warsaw). The intent is clearly to weaken Warsaw for Germany's turn 6. Germany probably should attack Warsaw on its turn 6, but ends up not doing it, I'm not entirely sure why, but I'm guessing for some reason the AI thinks moving troops towards Greece should be a higher priority than attacking Warsaw.
great_war-allies.tsvgSpeed is surely a priority for AI and I get the part of it quickly multiplying and you probably have to put limits to how complicated things the AI is allowed to consider. You might have to cut some corners, but I figure there's certainly room for improvement without slowing the AI down significantly.
-
@Quorthon4 That mostly just looks like bad luck as between Austria and Ottomans they should have killed on average at least 5 more units than they did. And then Germany would have had a strong attack opportunity.
-
@redrum Ok, so here are numbers I get with combat simulator (2000 simulation count):
Austria expected TUV swing: -49.62
Ottomans expected TUV swing: -17.22
on average both attacks should kill roughly 10 russian infantryGerman potential attack without the strafing attacks above expected TUV swing: -26.54
German potential attack with 10 less russian infantry to defend: 32.3
Total benefit from strafing attacks: -49.62 - 17.22 + (32.3 - (-26.54)) = -8 TUV swing -
@Quorthon4 Right but then Germany on average would get something like ~30 TUV swing and gain the factory. And ideally, you don't need to Ottoman attack. So its really -50TUV for Austria for 30 TUV for Germany plus they gain the factory.
I think its pretty close with average rolls. I don't think its a terrible gamble on strafing.
-
@redrum fair enough, there's also strafing attacks against same area round 8
-
@Quorthon4 Yeah, looks reasonable as again Austria gets poor luck though Germany is still able to at least take the factory.
-
@redrum assuming you value the factory that high, then yes. but I don't think it's worth ~ -27 TUV turn 5 or -60 TUV turn 8 to take the factory.
-
@Quorthon4 Yeah, maybe the AI overvalues the factory a bit. Though the turn 8 would be -60 then +30 so only -30 total on average to take the factory.
-
@redrum yeah, sorry, it's like you said closer to -30. Then you probably should compare that to the other attacks that wasn't made that could have potentially had positive TUV.
-
@Quorthon4 Yeah, these are both close calls on whether its worth it especially when you look at potentially what they could have gained otherwise. But I don't think either of them is terrible either. There is a pretty significant amount of randomness as well in the AI where they might choose other options if you replay the same game on those turns.
-
@redrum Thanks for the response, I overwrite save games as I play them, but I'll start trying to keep track of examples.
-
@redrum Also, I agree there is no "golden ratio" for any map/player/etc, but for any map/side/situation (of the maps I've played), the bigger the stack, the more it benefits from having a diversity of units. I agree someone would argue 15/4/2/2 vs 20/4/2/2 til they're blue in the face, but either of those armies will beat 25/0/0/0 every time.
-
@luhhlz Agree. The AI tries to buy a variety of units and tries to categorize them into types like fodder vs defensive vs offensive vs multi-move. It then uses these categories and checks the situation around a factory like whether its threatened or how far from the front it is to balance weights across these. Then it weights the units within each category (so infantry in NWO have a very high weight in the fodder category) and randomly selects units based on these weights.
So for say a factory near the front the AI on NWO will try to buy some fodder (usually infantry) and some defensive/offensive units (elite, artillery, etc) to give more strength. For factories further from the front, you'll see it buy more tanks, mech inf, mot inf, as it wants higher move units.
Trying to determine how much of each category and how to weight units within each category is tough as maps can have very different unit sets. The above is also somewhat simplified as there are a bunch of other factors as well but those are kind of the core factors.
-
In the enthusiasm for news of new tripleA WW1 map stuff, I've been watching the AI try to stalemate itself in that old Great War map I made with Surtur ages ago hehe. Currently skynet has bloodied itself into the 10th round. Going to just let it go tonight till it breaks itself or my memory. So far the HardAI seems to be managing fairly well for itself on either side. A dose of nostalgia from an oldy. For all its idiosyncrasies, it was fun map for the time, trying to bend the engine in early days hehe. But I haven't actually opened it up since 4.0, and now the AI is much much improved from back then.
Its been fun to watch the machine actually cobble together a vague kind of plan, compared to the nuts ass stuff it used to do with the AI when this map was first made like 10 years ago hehe. Cool to have an AI that gets the job done, pretty damn well actually, with something based on the old v2 rules.
CP are finally pushing where they can, but Allies just saw the US hop into the mix, so still a nail biter over here lol. I don't think the boys'll be home by Christmas this year. Nice work Hard AI
Great War Hard AI round 10.tsvg
Getting my kicks off for the era, watching the AI lamp itself before I crash into bed hehe. But will dream of badass 1914 PvP stuff to come for sure.
night all
Hello! It looks like you're interested in this conversation, but you don't have an account yet.
Getting fed up of having to scroll through the same posts each visit? When you register for an account, you'll always come back to exactly where you were before, and choose to be notified of new replies (either via email, or push notification). You'll also be able to save bookmarks and upvote posts to show your appreciation to other community members.
With your input, this post could be even better 💗
Register Login