Machine Accelerated Learning

Discussion in 'Intelligence & Machines' started by wesmorris, Jan 20, 2003.

Thread Status:
Not open for further replies.
  1. wesmorris Nerd Overlord - we(s):1 of N Valued Senior Member

    Messages:
    9,846
    I mean, okay.. if one can effectively model the human brain. Well, there are too many implications in terms of inducing false memories, etc. I'm curious though what effect you think it will have on education? I mean, if you can interface directly with the mind then "matrix" (like on the movie) kind of learning becomes possible right?

    What about having the subjectively conceptually indexed library of congress in your head? How long before this kind of thing? Does it put Universities out of bizness. Will "they" let it happen? Will "they" taint it horribly to the brainwashed demise of the human race? Is this a tired idea for a thread? What do you think?
     
  2. Google AdSense Guest Advertisement



    to hide all adverts.
  3. Adam §Þ@ç€ MØnk€¥ Registered Senior Member

    Messages:
    7,415
    Well, all that info going into our brains now is electrical signals. I have no doubt we will eventually be able to model the brain well enough to allow such accelerated learning. And no doubt the material being fed in will be as biased and tainted as that which we see in the press and in schools now. However, it will reduce the cost of education. Perhaps it will be a good thing, giving someone a university education by age ten or lower, for only a few hundred dollars, or downloaded free off the web. But then it would also be a great way to control the masses if one body was responsible for the information.
     
  4. Google AdSense Guest Advertisement



    to hide all adverts.
  5. wesmorris Nerd Overlord - we(s):1 of N Valued Senior Member

    Messages:
    9,846
    I've wondered for a long time now if it is in fact possible to present information without bias. The human mind seems to bias everything inherently.. and least in regards to history and the likes, scientific context. I don't know how you can unbias context I guess.
    No doubt, the potential is kind of disturbing. I wouldn't imagine it would take long before you reach max storage capacity for the human mind.... hmm.. so what then, just let the web cruise through your head.. instantaneous tap on the information flow of the world.. all through cell phone type of technology?
     
  6. Google AdSense Guest Advertisement



    to hide all adverts.
  7. Rick Valued Senior Member

    Messages:
    3,336
    What i dont really unerstand is How this can happen in such a small measure of time as shown in matrix...

    As in my previous threads i explained my Weights theory of Neural Networks(Not mine exactly

    Please Register or Log in to view the hidden image!

    ) so to actually have a stronger net connection i will have to test that electrical signal myself.If whatever was given as input works correctly a higher weightage is given to that meaning that the pattern of that signal fed into will become stronger. now this testing takes time.Subconsious mind will interact with Conious mind and then body will try to live through the experiences and then know.

    I however we are talking of forceful learning then my speculating is useless here.


    thanks.


    bye!
     
  8. hlreed Registered Senior Member

    Messages:
    245
    I have been working on brains for machines for a long time. So far the AI idea of learning has not been applicable. However I think there are two ways to learn in a machine.
    1. rewire.
    2. make use of a different combinatorial input
    A neuron has inputs and one output. The output of a neuron depends on the state of all its inputs. If you wrote down the numbers in each inputs for every possible input you find the combinatorial complexity of the neuron. There is a formula to compute that: Let S be the number of inputs. Let b be the base of the input- output system. Let C be the combinatorial number.
    C = b^S ; is the number. With two inputs and a 8 bit number
    C = 256. S = 8, C = 16777216 possible numbers.
    In life, normally only a small percentage of those numbers are seen. The neuron might see only 10 or 12 numbers during a day.
    But when a new situation arises, the neuron sees new numbers and produces new actions.
    Replicate this all over the brain, and there are a huge number of possibilities in even a small brain.
    So learning can occur simply by utilizing possibilities already there.
    That is my thinking so far.
    Movies are not real. Stories are not real. There is nothing there to make a judgment about learning.
     
  9. Rick Valued Senior Member

    Messages:
    3,336
    Hlreed,

    very interesting indeed.Thanks for your Input.

    Few confusions i would have loved to be clarified from you.

    • What is Combinatorial Complexity,exactly?
    • What is Base of Input and Output System of a Brain?
    • Replicate this all over the brain, and there are a huge number of possibilities in even a small brain. -->the above quoted statement of yours has a problem.How do you define all over the brain,By counting Neurons?By Speculating?By finding the UpperBounds or Lower Bounds for a Neuron Numbers?

    help would be appreciated.

    thanks.
    bye!
     
  10. hlreed Registered Senior Member

    Messages:
    245
    Think of the axon of the neuron as a digit. Then think of all the digits in the brain as a number. This number changes every instant, but it traverses a path.
    The combinatorial part comes about when you try to list all the possible inputs to a Haltree or neuron. If b is the number base and S is the number of inputs, then C = b^S is the number of possible inputs. Here is a binary count of 3 inputs.
    number/counts
    0 0 0 0
    0 0 1 1
    0 1 0 2
    0 1 1 3
    1 0 0 4
    1 0 1 5
    1 1 0 6
    1 1 1 7

    2^3 = 8
    One note. The output of a tree must have the same base as its inputs. What happens here is the output is b but it is repeated
    C-b times, so in context with all the other digits, it still hase many meanings.
    There is a lot to this. I left you my web site on another thread.
    Check it out. I will be glad to teach all of this to you.
     
Thread Status:
Not open for further replies.

Share This Page