@thedog You may want to submit this as an actual issue with AI log. It maybe that it always focusing on the big opposing armies and view the neutral spaces as low priority.
Recent Posts
-
@dsiof nice! can’t wait to see what you come up with
-
@dsiof In how far is this a bug report? Just asking because I wonder why you chose "Bug Reports" to post this...
I have moved this topic to 'AI' for now. Maybe it would be useful to divide this topic into its two subjects: AI and Calculator.
Anyway... good luck!
-
Hello, I've already posted in the axis and allies forum, but since the platform for the GNN I'm trying to build is here on TripleA, I was advised to post here as well.
I'm trying to build a neural network for a couple of Axis & Allies maps, specifically starting with the simplest 1941 game, then going to one of the three of 1942, Anniversary 40th, and G40, depending on how well we do with '41. It seems others recently have already been doing something similar, so hopefully there are people who can help. Me and a friend are trying to make a semi-competent AI that will hopefully improve and be better than the AI already in the game.
We are currently doing another project, where we are trying to find a simple formula to calculate battle odds without a battle calculator for in-person play, with hopefully about 5 variables but may be more or less depending on accuracy. Any advice for either is very helpful, currently we only have developed a pretty accurate 1941 battle calculator, but are fixing a bug where in naval battles with battleships the odds are 1% lower than the 1942 calculator (the most accurate seeming calculator we could find) had stated, but only had an error range of +-.4%. We will also try the TripleA battle calc, as previously we had used David Skelly's calculator but that seemed to be a little less accurate and consistent as 1942.
