I know current learning models work a little like neurons but why not just make a sim that works exactly like how we understand neurons work

  • x86x87@lemmy.one
    link
    fedilink
    arrow-up
    82
    ·
    6 months ago

    Simulating even one neuron is very complex. Neurons in artificial neuron nets used in machine learning are a gross oversimplification. On top on this you need to get the wiring right. On top on this you need to get the sensorial system right (a brain without input is worthless). On top of this you need an environment. So it’s multiple layers of complexity that we don’t have

    • BastingChemina@slrpnk.net
      link
      fedilink
      arrow-up
      6
      ·
      6 months ago

      What I find fascinating is the efficiency of the brain.

      With a supercomputer and the energy of a nuclear station to run it we are able to simulate a handful of neurons interacting with each other.

      On the other hand the brain with billions of neurons only requires the energy of one or two potato to run.

      • x86x87@lemmy.one
        link
        fedilink
        arrow-up
        3
        ·
        6 months ago

        To be fair, nature had millions od years to optimize the power consumption and we only observe the successful results since the failures didn’t survive.

      • Brickardo@feddit.nl
        link
        fedilink
        arrow-up
        1
        ·
        edit-2
        6 months ago

        We’re having our particular technological revolutions as well. In little more than a century we’ve managed to construct computing devices with capabilities that may have taken thousands of years to be achieved by nature.