A few questions about AI of RISK.

AI discussion, ideas, and SDK help.
Post Reply
User avatar
quartaela
Luxer
Posts: 14
Joined: Tue Jun 25, 2013 1:52 pm

A few questions about AI of RISK.

Post by quartaela » Wed Jun 26, 2013 5:49 am

Hi everyone! I want to ask a couple of questions about AI of RISK. First of all i am very new on AI topic and have been studying it for a few months and i want to implement RISK game. Well i have an idea about minimax, expectimax algorithms for implementing a decision tree and i'm sure that i have to use expectimax algorithm. However, i'm unsure about should i use A* algorithm for finding closest path from a territory to another ? or is it mandatory ?. I am asking cause there is a very useful document written by
Wolf Michael - "An Intelligent Artificial Player for the Game of Risk
and it says
It also offers a method that utilises an A*-Search [Russell and Norvig, 1995] to compute the attack path with the highest victory probability given a start territory and a destination territory.
. I have a some opinions but didnt consider to use A*. So if you can help or share your opinions i will be very appreciated.

User avatar
dustin
Lux Creator
Lux Creator
Posts: 10433
Joined: Thu May 15, 2003 2:01 am
Location: Cascadia
Contact:

Post by dustin » Wed Jun 26, 2013 6:54 pm

A* is definitely a useful algorithm to use.

You should check out the Sillysoft SDK, which also has source code for a number of AIs, and helper functions for path-findings as well:
http://sillysoft.net/sdk/

User avatar
quartaela
Luxer
Posts: 14
Joined: Tue Jun 25, 2013 1:52 pm

Post by quartaela » Thu Jun 27, 2013 1:21 pm

thanks dustin. i was just curious about that if A* is used for pathfing in ai. well i guess i have to spend time on trying to understand the mechanism of lux agents. :)

Post Reply