Re: What did that thread indicate?

On 26 Sep 2005 01:07:04 GMT, curt@xxxxxxxx (Curt Welch) wrote:

>Your system has always struck me as overly complex. I don't see anything
>you are doing in your 6 or so modules that I'm not doing in one network.
>But I know that you also think I'm missing 5 or so modules. :)

There is a need for at least 6 integrated subnetworks, each with its
own specific function and principle. If your system is a single huge
network, you're doing something wrong. Why? Because you're not seeing
the whole problem. You may want to believe otherwise but your
understanding is lacking.

Kurt, every time I mention a problem, you reply that your network
already does it or that it is about to do it. If your network was so
advanced and you really had the correct solution to the problems I
mention, you would not be wasting time on usenet. You would have
already found the solution to AI and you would be well on your way to
become the richest man on earth. From what I read from the articles
you post here, you're not even close to a solution. Although, I must
say that you're orders of magnitude closer than the GOFAI crowd. One
of the things that you don't yet appreciate is the principle of
complementarity (yin and yang). It's the most important principle of
them all. Without it, you might as well be sucking a mule's hind tit
because you won't get very far. Yet, you never even mention it.

>All my conflict resolution is built into the design of my network by the
>fact I never split a signal. I started with a foundation where conflict
>was impossible so that prevented me from having to build active "conflict
>resolution" technology into the design. I suspect the brain is doing what
>you do - solving it by fixing the problem after the fact with a conflict
>resolution systems.

It cannot possibly work that way and, besides, as you say, that is not
how it works in the brain. There is a huge number of things being
processed in parallel and they generate motor output. There are only
enough effectors to do a limited number of actions simultaneously.
This is part of the attention problem. If your network does not
process many signals in parallel and, as a consequence, is forced to
resolve conflicts as it learns, you are doing something wrong.

>And at the low level, the correct way to pick behaviors, is not to actually
>predict what will happen in the future, and do some sort of silly
>computational intensive "search" of options, but instead, to build a value
>prediction system which evalutes the worth of different behaviors, so the
>low level hardware is always selecting the best behavior now based on it's
>knowlege about past experience with this behavior.
>
>So, as I asked above, what type of prediction of the future does your
>hardware make and what does it do with it's prediction?

When we are thinking, we are constantly evaluating the outcome of
behaviors. That's what prediction is for. To evaluate outcomes, we
must be able to follow a train of thought (a sequence) to see where it
leads. Without the ability to anticipate, we would not survive.

In addition, without anticipation, it would be almost impossible to
understand the environment. Why? Because more often than not, objects
are partially occluded from view, and sounds (especially speech) are
either partially unintelligible or masked by other sounds. We must be
able to fill in the gaps. In the business, this is called pattern
completion. Read also about the cocktail party problem. It is
impossible to follow a conversation in a noisy environment without an
anticipatory mechanism.

And please, Kurt, don't reply that your network already does that. It
does not.

[cut]

>> >The solution was found with the design of my current network.
>>
>> [cut]
>>
>> Kurt, I read your post and I don't believe you have the solution.
>
>Yeah, that's because we all believe we are on the right, and only path, so
>anyone on a different path is just missing something. :)

Yep. But some do understand more than others.

Louis Savain

Why Software Is Bad and What We Can Do to Fix It:
http://www.rebelscience.org/Cosas/Reliability.htm
.