On 2/23/15, Jim Bromer via AGI <[email protected]> wrote: > We programmers can become habituated in our thinking about mundane > things like variables and procedural methods in programming. This > thinking would not be a problem if we were able to keep an open mind > when we need to develop ideas in special ways but the problem is that > a flexibility of mind tends to become weaker over time just because > our typical mindsets work so well in normal circumstances. > > However, when working on problems for which conventional and standard > methods have failed, then it is definitely time to rethink your > assumptions about the tools that you use. > > Many of the things that I could tell you about might seem trivial > because you probably already have many of the basics down. The > problems come from implementation errors (of thought) that you don't > see coming. Just because a standard technique (of thinking about > something) is usually integrated into other related (thought) objects > in certain ways that does not mean that is the only way that those > thought processes can be used. You know that of course. The problem is > that you don't know how to find a new implementation of integration > rules in order to deal with a particular problem. It is similar to the > AGI Frame Problem as I conceive it. When dealing with a new kind of > problem you have so many possibilities that you can work with it > becomes practically impossible to find the right tool (or the right > knowledge) for the job. What I am trying to say is that these tools of > mind may be comprised of component systems just like the components of > programming constructs. The problem is that your thinking can become > so standardized by the habitual use of these programming constructs > that it can become almost impossible to think about them in different > ways.
Jim, have you taken a look at some of the more far out (seeming) newer architectures? eg., http://rala.cba.mit.edu/ (I think this is the one I was intending, it's getting a bit late here....) I guess mostly I think in terms of neural nets or symbolic, but it helps to dive into some of these architectures. Mike A ------------------------------------------- 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
