He mentioned that the program should make simple decisions and limit information under consideration. These are two aspects of using componential knowledge. (I don't know what it should be called). Another reason to use componential knowledge is that it provides us with possible insights on how the program should use new knowledge based on the categorical similarities of previously learned knowledge. This kind of insight has the potential to avoid the AGI pits of strict hierarchal knowledge and poorly integrated knowledge. However, this is not the sort of thing that is currently practical for game AI. A computer game has a lot of concrete knowledge of objects and events which the Game AI programmer has to use to make his AI work, but it does not lead to the solution of AGI problems. Another reason why componential knowledge is good is that the alternative is impossible. Jim Bromer Jim Bromer
On Thu, Jan 16, 2014 at 11:54 AM, Alan Grimes <[email protected]> wrote: > Here's an article I just skimmed about game AI. > > http://gamasutra.com/blogs/DavidParis/20140113/208505/Gameplay_Programming_Hints_Building_AI.php > > > -- > Powers are not rights. > > > > ------------------------------------------- > AGI > Archives: https://www.listbox.com/member/archive/303/=now > RSS Feed: https://www.listbox.com/member/archive/rss/303/24379807-f5817f28 > Modify Your Subscription: > https://www.listbox.com/member/?&; > Powered by Listbox: http://www.listbox.com ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657 Powered by Listbox: http://www.listbox.com
