The Powder Toy

TPT evolution? by perogiepro

perogiepro
perogiepro
99 / 13
24th Aug 2015
22nd Dec 2015
We are the SFPI, we are searching for mechanical life that can detect danger and think. essentialy basic AI. Our end goal is to make one that can replicate and change, so it will """evolve"""
machine intelligence experiment neuron life survival challange evolution spfi

Comments

  • nlatham
    nlatham     25th Aug 2015
    you could use guillians pistin pulsars
  • perogiepro
    perogiepro     25th Aug 2015
    @msasterisk your neural network is perfect!. im going to fit it into this save.
  • msasterisk
    msasterisk     24th Aug 2015
    Thanks, classeh, but the eye isn't the problem. Its just that the robot for some reason can't handle the signals, even though they go through a depressor... but like I said, thanks!
  • classeh
    classeh     24th Aug 2015
    check out my organism for eye, it's big but it works
  • msasterisk
    msasterisk     24th Aug 2015
    Wowwww I attached my neural net to the random robot. I noticed that it was following a repeating sequence- but the seqquence changed a little each time! EVOLUTION
  • msasterisk
    msasterisk     24th Aug 2015
    ID:1842276 It's a large brain with a robot attached. So far the eyes are nonfunctional though.
  • msasterisk
    msasterisk     24th Aug 2015
    Heyyyy I did the same thing- but instead of trying to evolve my robot I kept its environment perfectly clean and the robot intact. This is a way more interesting way of looking at it though!
  • Schmolendevice
    Schmolendevice     24th Aug 2015
    @perogiepro Sorry for clouding the comments with text, but that mini 'intelligent' robot looks wonderful as it squibbles around! I'd actually be delighted to use my own neurons and incorporate inhibitors that ensure only one signal is ever going into the robot at a time.
  • Schmolendevice
    Schmolendevice     24th Aug 2015
    id:1844423 These are the building blocks I plan on using. In ways, larger 'HSWC somas' will be harder to 'excite.' After that you can add as many synapses and 'TSNS axons' as you like and connect them together. Think of it like this, stimulus comes in and eventually goes out to some part of the body but can be excited or inhibited. It is our job to design what patterns of stimulus may 'encourage or discourage' that stimulus from causing a _response_.
  • Schmolendevice
    Schmolendevice     24th Aug 2015
    In 'some time' I'll be posting a save on how to make arbitrary neural systems on your own however you want where _we_ as TPT users become the driving force in this 'evolutionary contest.' We are the 'mutative force' who prune different designs through our own community facilitated natural selection. Single minds may be able to evolve, but physical evolutions will have to happen between saves.