Steve, John, Mike,
thanks for all the comments. There are two different things on the table. Plan A is the traditional well-known approach. You learn about what you think is important and write a computer program about that. In this approach, we use our brains to create associations in the knowledge, bind it, and obtain the structures that effect the self-organization. We can even create rules for doing some bindings and put the rules on the computer. For example, to obtain self-organizing agents, as John describes. Plan A has the advantages we all know and the limitations we also know. Plan B. There is a process of inference in the brain that self-organizes knowledge by creating associations. I have discovered how it works, and it is actually very simple. I don't know how the brain implements the inference, but I have reason to believe that the inference is widely distributed rather than being concentrated in any one particular place. The limitation of plan A is that, when we use the inference in our brains to create the algorithm, and then copy the algorithm to a computer, the inference stays in the brain and doesn't get transferred to the computer. The computer remains as incapable of doing self-organization as it was in the first place. So plan B is to install the inference on a computer, and then one will have no need for the brain anymore. >From the point of view of the user, the difference between A and B is actually tiny. In A, you learn all you can about something, apply the inference yourself, and force-feed that to the computer. In B, you learn all you can about something, and you give it directly to the computer, as it is, raw, unprocessed. Then the inference in the computer will self-organize the knowledge, and create the same algorithm as before. Or, perhaps, you just let the computer itself to learn, for example from a sensor. In A, knowledge is program. In B, knowledge is input. Which one looks closer to how people work? I hope I have not offended some people, most notably Alan, by sounding as if I disdained their engineering skills. I was only trying to send a very strong message that knowledge is only input in plan B, and can not be used as program. You must have zillions of concerns. I'll do what I can to answer. But consider how difficult my position is. Multi-disciplinary communication is very difficult. Suppose I say: "Emergent inference has a very strong foundation in theoretical Physics." The word strong does not really mean much to you unless you know exactly what you would be missing if you ignored my statement. And you don't know that unless you have an extensive knowledge of physics. What can I do? I can't just shut up and dissapear in a hole. I intensely feel the pain of those who ruined their careers by working on plan A, and I just as intensely feel that I am right and the others are wrong. There are many other things I can do that I haven't done yet, but they all take time and I need help, I want others to share the burden so we can go faster. If not, I'll just keep going at my own pace. Sergio Here is a metaphor I thought might help to see the difference between A and B. In the past, if you had a vessel and you wanted to transport some cargo, you would just buy yourself some slaves and make them row. Today, if you have a computer and wanted some task done, you hire some developers and make them write code. Slaves do not teach a vessel how to row. Developers do not teach a computer how to develop. AGI | <https://www.listbox.com/member/archive/303/=now> Archives <https://www.listbox.com/member/archive/rss/303/18883996-f0d58d57> | <https://www.listbox.com/member/?&; ad2> Modify Your Subscription <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-c97d2393 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-2484a968 Powered by Listbox: http://www.listbox.com
