If my impression of these discussions > is accurate, if the partisan arguments for logic, probability or > neural networks and the like are really arguments for choosing one or > the other as a preponderant decision process, then it is my opinion > that the discussants are missing the major problem. That issue is the > way ideas interact. > > Isn't it true that your advocacy for probabilistic methods is an > argument for using probability as a primary decision process?
I don't really know what you mean by "a primary decision process"? In OpenCogPrime, probabilistic logic is not " *the* primary decision process" However, it is one of the major ingredients that goes into decisions the system makes (according to the design... note that PLN is not fully integrated into OpenCog yet, though we've played with it in the NCE) Other major ingredients include various heuristics for concept creation/ interaction, including e.g. conceptual blending which is key in the cognitive theory of creativity... The action selection process makes use of inputs from concept creation, probabilistic logic, and other ingredients. So again I'm not sure how to shoehorn my AGI designs into the categories your terminology implies... -- Ben G ------------------------------------------- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244&id_secret=114414975-3c8e69 Powered by Listbox: http://www.listbox.com
