Search For Powder Intelligence by Windspren

Windspren

30 / 4

2nd Jul 2016

29th Sep 2016

The reborn SFPI! Our goal is to develop and identify forms of pseudointelligence and life within TPT in the forms of PSTN bots, computing systems, genome emulators, brain architecture simulations, and anything else we (or TPT AIs we find :D) can think of

sfpi tpte3

Comments

• 

  QuanTech 3rd Jul 2016
  sure! i'll join!
• 

  Windspren 3rd Jul 2016
  Do you guys want to join this? ID:2009917 I'm planning on making SFPI a subdivision of it.
• 

  QuanTech 3rd Jul 2016
  oh, that is very nice! as well, you can fit more TUNG particles than mercury injectors! thanks for the idea!
• 

  Windspren 3rd Jul 2016
  Yes, brain developers have been using that concept for a while, but instead of a MERC injecter, we usually make portions of the circuit touching the MERC out of TUNG, or mix in NBLE with the MERC. The heat produced from current flowing through those materials makes the MERC expand, allowing a brain to "learn" by making it more likely that current flows through a certain neuron every time it is SPRKed.
• 

  QuanTech 3rd Jul 2016
  i made a (poor) theoretical model for a perceptron. So far,i have only added the weight. Here is the id: 2010283
• 

  QuanTech 3rd Jul 2016
  i think one way you can implement perceptrons is to make a controlled mercury injector, which will act as the wieght, so that the more mercury there is, the more probable the output is the regular input! (e.g input 1 with weight 10px mercury = high chance of output 1)
• 

  QuanTech 3rd Jul 2016
  if you haven't heard of pereptrons, they are basically a basic model for neurons. In the simplest example, there are 2 inputs, 1 regular input (from other calculations) and 1 weight for the input. The weight is the priority of the input, so highly weighted inputs will most likely cause the perceptron to output that input. Sorry for the terrible explanation, just google it if you do not know what they are!
• 

  QuanTech 3rd Jul 2016
  oooh nice! Have you heard of perceptrons? This seems very similar to that!
• 

  Windspren 3rd Jul 2016
  ID:2010250
• 

  Windspren 3rd Jul 2016
  What I've done so far, as a byproduct of my natural selection test.