Difference between revisions of "SOC2014 Daniwa Ai"

From The Battle for Wesnoth Wiki
(Description)
(Description)
Line 9: Line 9:
 
'''Abstract:''' Implement a process that would allow the AI to make correct decisions in regards to attack and retreat. It will be based off probability focused on current metagame habits.  
 
'''Abstract:''' Implement a process that would allow the AI to make correct decisions in regards to attack and retreat. It will be based off probability focused on current metagame habits.  
  
'''Proposal:''' The focus of my implementation is not on outputting the optimal result in every case but instead will aim at mirroring a player like decision. The reason for this is to foster and develop good game decision making in new players and prevent bad habits. Ideally, the AI should act correctly but unpredictably (which is technically impossible) and be able to be a training aid.
+
'''Proposal:''' The focus of my implementation is not on outputting the optimal result in every case but instead will aim at mirroring a player like decision. The goal for this is to foster and develop good game decision making in new players and prevent bad habits. Ideally, the AI should act correctly but unpredictably (which is technically impossible) and be able to be a training aid.
 
 
'''TODO:''' Set the proper Category in the top of the page (<nowiki>[[Category:XYZ]]</nowiki>), to match the one you'd like to apply for, from the list on the main summer of code page (e.g.: <nowiki>[[Category:SoC_Ideas_AI_Global_strategy]]</nowiki>). Otherwise your idea will be marked as a new one.
 
  
 
==IRC==
 
==IRC==

Revision as of 16:49, 18 March 2014


This page is related to Summer of Code 2014
See the list of Summer of Code 2014 Ideas



This is a Summer of Code 2014 student page


ATTENTION

  • This page is under construction.

Description

Improve the AI's Attack and Retreat decisions.

Abstract: Implement a process that would allow the AI to make correct decisions in regards to attack and retreat. It will be based off probability focused on current metagame habits.

Proposal: The focus of my implementation is not on outputting the optimal result in every case but instead will aim at mirroring a player like decision. The goal for this is to foster and develop good game decision making in new players and prevent bad habits. Ideally, the AI should act correctly but unpredictably (which is technically impossible) and be able to be a training aid.

IRC

TODO: Put only your IRC nickname in there. In case of alternate nicks, separate with commas.

Example contents for this section (without the quotes): "user" "user1, user2"

Questionnaire

TODO: fill out the questionnaire, on your copy of this page. The questions are provided in the SoC Information for Google page.