How can I influence the minimax algorithm to prefer immediate rewards?

Question

I use minimax for the game Strategyo (where the computer has full knowledge of all parts). However, I find that the computer often does not attack a piece that it can easily destroy. From what I understand, minimax scores come from leaf nodes of the displacement tree (where each level is a turn, and each point for the leaf node is calculated using the score function for the board in this position). Therefore, if I have a depth of 3 levels, the computer can choose an attack when moving 1 or attack when moving 3. According to the minimax algorithm, it has the same score that is associated with it (the resulting position on the board has the same score). So, how do I influence the minimax algorithm to prefer immediate rewards for possible rewards? those. I would like the rating to fall over time,but with minimal work, I don’t see how this is possible. Minimax always uses leaf nodes to define intermediate nodes.

Answer 1

As others mentioned in the comments, minimax should be able to notice if there is a danger in delaying the capture of the piece automatically, and changing the rating function to make it prefer the previous captures is likely to be detrimental to performance,

, , , - ( ). , ( ) . , , .

, ; 2 4 3. , , , . , .

, , . , ( , , ).