Improved Pro AI, please give it a try and post feedback

[Cernel]

@LaFayette This is the game-save:
2026-11-1-World-War-II-v5-1942-SE-TR.tsvg

I'll also point out that, for the remaining part, the game-play is not very good: the AI is taking too much risk and not focusing enough. It should have sent everything (EVERYTHING) it can send to Ukraine SSR and everything else or almost everything else which can reach West Russia to West Russia.

I guess (once the max is sent to Ukraine SSR) sending 2 infantries and 1 artillery to Baltic States and all the rest to West Russia would also be acceptable, but I'd say better only attacking Ukraine with everything possible and West Russia with next everything possible.

In any case, for sure Ukraine SSR must be attacked with max possible. That is out of doubt.

[LaFayette]

I'd love to see extended write ups even. A save game with every blunder pointed out, every flawed strategy, and any praise for what the AI did would be perfect. The analysis step of the write ups is almost entirely automated.

Those write ups would lead to quick and substantial improvements to AI play.

The AI play against itself leads to very small improvements, usually bug fixes and not deep strategy improvements. These human vs AI games and detailed write up is (at the moment) by far the best way to further improve the AI. Since the analysis step is largely automated, if 50 writes up landed today, the AI available tomorrow would likely be several echelon's stronger.

[Cernel]

@LaFayette About the attack in Ukraine SSR, the only actual difference is that both fighters would go there (as part of sending there everything which can reach). The AI actually sent everything at reach except 1 fighter (which was instead sent to Baltic States).

[LaFayette]

Perfect, thank you @cernel ; keep the feedback coming!

PS: favor 'edit' to add additional notes on the same save game, so that one post and one save game match together. It keeps the analysis a little bit easier.

[LaFayette]

@cernel a patch is being built - check the latest in about 10 minutes from now.

edit: patch is built and available (Hard Beta AI)

https://github.com/triplea-game/triplea/pull/15159

The algorithm change identified:

• the Pro AI was sending spare fighters first to the lowest odds battle
• the update, Pro AI now sends excess fighters to the highest value battles first that are not yet certain wins

So, Pro AI will now ensure the high value battles first. This update changes the behavior to generally send figs as you described 90% of the time.

[LaFayette]

@cernel - Sub behavior now updated from: 2026-November-01 - 2026-11.16.15866

The sub should be moved 100% of the time now.
The previous release had the fighter fix movement fix.

[Cernel]

@LaFayette I have not tested the new one yet.

I don't know if this may be of help, but I wanted to make clear why attacking only Ukraine S.S.R. and West Russia the way I described is better than what the AI did.

In the latest game-save which I've posted, the AI makes 3 attacks.

territory: forces sent: average TUV swing + %Win * territory production value = average TUV swing + average production increase = total bonus
Ukraine S.S.R.: 3 infantry, 1 artillery, 3 armour, 1 fighter: 12 + 71% * 2 = 12 + 1 = 13
West Russia: 5 infantry, 2 artillery, 1 armour: 8 + 94% * 2 = 8 + 2 = 10
Baltic States: 3 infantry, 1 fighter: 3 + 89% * 2 = 3 + 2 = 5
TOTAL: : : 23 + 5 = 28

Now, the first most obvious mistake we can see is that 1 infantry is left in Karelia S.S.R. during Combat Move.
That infantry could have gone to Baltic States or to West Russia to attack.
Baltic States is questionable because it is highly subjected to counter-attack.
West Russia is not highly subjected to counterattack regarding the average result with the infantry added and the territory conquered (which is 6 units left if the attacker won, specifically 3 infantry, 2 artillery and 1 armour). Assuming Ukraine S.S.R. taken, the Germans on the next turn can send to West Russia up to 3 infantry, 1 armour, 5 fighters. The battle between these forces and 3 infantry, 2 artillery and 1 armour is 8 TUV swing, which is much less than what those forces can do if used elsewhere, which means the relative risk is virtually inexistent.

These are the statistics same as above but sending 1 more infantry (from Karelia S.S.R.) to West Russia

territory: forces sent: average TUV swing + %Win * territory production value = average TUV swing + average production increase = total bonus
Ukraine S.S.R.: 3 infantry, 1 artillery, 3 armour, 1 fighter: 12 + 71% * 2 = 12 + 1 = 13
West Russia: 6 infantry, 2 artillery, 1 armour: 9 + 98% * 2 = 9 + 2 = 10
Baltic States: 3 infantry, 1 fighter: 3 + 89% * 2 = 3 + 2 = 5
TOTAL: : : 24 + 5 = 29

Instead, sending everything which can reach to Ukraine S.S.R. and the everything remaining which can reach to West Russia is the following.

territory: forces sent: average TUV swing + %Win * territory production value = average TUV swing + average production increase = total bonus
Ukraine S.S.R.: 3 infantry, 1 artillery, 3 armour, 2 fighter: 18 + 89% * 2 = 18 + 2 = 20
West Russia: 9 infantry, 2 artillery, 1 armour: 10 + 100% * 2 = 10 + 2 = 12
TOTAL: : : 28 + 4 = 32

This is a total of 5 average TUV swing more and 1 average production increase less for a grand total of 4 more than what at the game-save.

This is per se only a small difference, but it is something.

The other main element is that sending to Baltic States is just not worth it because there are many Germans units which can attack there and nowhere else, so that is a very easy counter-attack for Germans with almost certainly no counter-attack for Russians thereafter. That may be worth it if the Russians would have had spare air units, but this is certainly not the case as they are all required in Ukraine S.S.R..

[LaFayette]

@Cernel Useful feedback, I'm toying around with the idea of "don't create targets for your opponent". A first metric I'm thinking of is "how much firepower could the opponent bring to bear". That's a good single sum. It's kinda complex though how to weight that signal. I'm instrumenting the AI (for benchmarking purposes) at least to capture some data around that metric and when it is clearly useful to consider.

[Schulz]

I tested the Allies Hard AI.

My game has a special rule allowing fighters to scramble to sea zones without airfields. However, the British tried to attack the Japanese bomber on Borneo with a lone transport, resulting in the Japanese fighters sinking the British transport during the scramble phase.

They also left two undefended transports in range of the German bomber.

The battle screen froze. Also, I would like the default battle screen to be smaller if possible.

[LaFayette]

@Schulz would you mind attaching a save game for the scramble case. I'll try to look into the others.

[Schulz]

@LaFayette 2026-11-2-Axis-Ascension.tsvg

[Cernel]

@LaFayette It has actually more to do with the fact that the Russians TUV is the most valuable. Since Russians are the primary Axis target and are semi-isolated from the rest of the Allies, there is the dynamic by which the same Russians TUV is worth more than the same Germans TUV, which in turn is worth more than the same British TUV, which in turn is worth more than the same Americans TUV.

For example (I'm not saying that these numbers are correct.), the AI would need to understand that the Germans TUV is worth 90% the Russians TUV so giving a value to any TUV swing between Germans and Russians valuing the Germans TUV at 90% the Russians TUV, so Russians would end up on par as a matter of TUV if they destroy 10 Germans TUV for every 9 Russians TUV which they lose (destroying the same amount being a bad result for Russians).

I don't know if the AI can maybe be made to understand in general term who is the besieger and who is the besieged at the moment (the besieger being Germans and the besieged Russians at least early on) and so try to be TUV conservative with Russians and try to induce more reciprocal TUV destruction with Germans.

The other element is that in TUV destruction over multiple rounds via territory trading the one with the most air units is advantaged, and again here there is a major superiority of the Germans at least for some rounds (the Germans having many more fighters than the Russians, though of course the Germans fighters are distracted by the other allies much more than the Russians fighters are by Japanese).

For context, I've chosen "World War II v5 1942 SE TR" not because it's a map I know: I've actually never played this map against a human opponent. The reason is that this is the only extremely well balanced map which TripleA has amongst the basic ones. (Notice that this applies to this game but not to the original "World War II v5 1942 Second Edition", which is unbalanced in favour of Axis.)

[LaFayette]

@Cernel I think I have a patch that might be ready in a week that captures this. Essentially the concept of a "Front" as pairing any two clusters of factories. That defines attack paths along which attacking/defending is computed based on how many turns to overwhelm along that path. If there is no chance to overwhelm, then it's a defensive front vs offensive. From there, the income ratios comes into play. Your rule is a bit simpler, but does not generalize very well. If we can determine how to calculate those ratios, we would have a more generic way to establish that.

[beelee]

@LaFayette

Do you wanna add a ping to the Lobby Intro ? New AI to try post results here type thing ?

Idk

I imagine most Lobby Players blow right by the intro but but even one hit would provide data

I might ping something at the other site too. Seems Revised would be the main game or SE I guess. Idk as I don't play them.

I'm guessining it still doesn't do Politics so Anniversary and G 40 probably won't work but the Objectives definitely a good step.

Hmm ... Maybe Anniversary doesn't have politics ... just Global

Well at any rate I might post something under the triplea thread over there.

Maybe @panther has a suggestion for promoting it ?

Good Work Lafayette

[Cernel]

@LaFayette OK.

As an example of Russians being (obviously, as in every V-series game) the main focus of Axis attention, I've played the game going almost full on Russia and took Moscow round 5 against Hard Beta AI as of version 2026-11.16.15866. (I actually reloaded the game twice in favour of Russians until they had a good round 1 open, but they always played it perfectly: the first two times they just got bad dice. )
2026-11-2-World-War-II-v5-1942-SE-TR.tsvg

If you take a look at it, is there anything particular I should focus on commenting, or is it all fairly obvious to you by looking through history?

[LaFayette]

@Cernel I'd like you to comment. You have a better feel of what was significant. Second, it's a big time saver. Your comment and save game feeds an AI (Claude) analysis that can patch code, replay the save game, create data sets that the (TripleA) AI can be scored on.

I'm setting it up so that the AI is tested on a variety of different game scenarios. Blunders that it makes, rules it does not understand, and then (TripleA) AI improvements are tested through this battery as well as scored on overall ability.

A really thorough description/summary of the game and key insights is very useful here. If we can get 20 or so of them, that'd be huge. Going through the exercise personally, the playtime took me about 3 hours the write up maybe 10 minutes. Plus, I only know a couple maps deeply, which limits me further.

Processing of feedback is highly automated right now, it's pretty bad ass. I was able to parse out every complaint of the AI, and that was a big source of fixes/improvements.

[RogerCooper]

I tried the new AI. It seems a bit more creative in its play and handles naval matters better.

[beelee]

Well I posted at the other site lol

https://www.axisandallies.org/forums/topic/42759/major-ai-computer-player-update-to-triplea

[LaFayette]

@beelee Thanks! One nitpick, the latest is also (always) available at the main website: https://triplea-game.org/download/

[beelee]

@LaFayette

I updated with the main triplea zone. The one with the download triplea