Yet another possible way to go about ai

  • 3 Replies
  • 352 Views
*

ranch vermin

  • Not much time left.
  • Terminator
  • *********
  • 958
  • Its nearly time!
Yet another possible way to go about ai
« on: July 14, 2018, 10:11:32 am »
So Id like you consider a perceptron as an "approximating search engine."...  but in this one, the perceptron is looking inside another model, which is provided by the environment. (excuse me if youve heard that one plenty of times, im trying to do something slightly original here.)

So these perceptrons, as they evolve, hillclimb, or anneal or WHATEVER,  are maximizing a score function, which itself is hard coded, fixed in the system.

When your looking in this "model generated from the environment" the first thing something thinks of is,  "oh im developing an activity, or some reaction for it to do"  but this is actually only thinking about it in a boring obvious way!

These hillclimbing perceptrons could do lots more than just this!

Perhaps you could score generating 3d from 2d!   and after that you could clear up alot of object distinctions and really neaten up your system before you go develop the action.

So theres alot fine tuning that some prior reorganization could help you with,   but it could just be scoring functions the same!

*

LOCKSUIT

  • Emerged from nothing
  • Trusty Member
  • *************
  • Transformer
  • *
  • 2104
  • First it wiggles, then it is rewarded.
    • Enter Lair
Re: Yet another possible way to go about ai
« Reply #1 on: July 14, 2018, 11:00:33 am »
So it does backprop with a cost and finds the optimal actions for each given events?

I'm not sure if you should give us more details, or summarize your OP into a smaller point :P Try one or both though!
Emergent

*

8pla.net

  • Trusty Member
  • **********
  • Millennium Man
  • *
  • 1035
    • 8pla.net
Re: Yet another possible way to go about ai
« Reply #2 on: July 16, 2018, 06:15:29 am »
Multi Layer Perceptron

http://aihax.com/MLP/
My Very Enormous Monster Just Stopped Using Nine

*

ranch vermin

  • Not much time left.
  • Terminator
  • *********
  • 958
  • Its nearly time!
Re: Yet another possible way to go about ai
« Reply #3 on: July 16, 2018, 07:09:52 am »
its a form of "generate and test" isnt it.

Thats why neural networks can have a go at brute forcing ANYTHING,   but in some cases they dont do so well,  (they have problems with a needle in a haystack) but you really get imaginitive with how your going to solve your problem.

Thats all this thread really means,  that when your writing a network,  you can get a neural net and have a go at anything u dont want to write directly!

 


What kind of abilities will AGI have?
by LOCKSUIT (General AI Discussion)
Today at 02:21:40 pm
ETHICS
by Hopefully Something (General AI Discussion)
Today at 08:12:23 am
A big Hi to all bot lovers and makers!
by DemonRaven (New Users Please Post Here)
October 13, 2018, 09:08:26 pm
BOOST
by Art (General Chat)
October 13, 2018, 01:59:09 pm
collaborators wanted.
by DemonRaven (AI Programming)
October 13, 2018, 12:16:52 pm
XKCD Comic : Rock Wall
by Tyler (XKCD Comic)
October 13, 2018, 12:00:36 pm
How many lines of code will AGI be?
by Hopefully Something (General AI Discussion)
October 12, 2018, 11:16:25 pm
ChatbotML on Twitch
by squarebear (Home Made Robots)
October 12, 2018, 09:26:00 am
Tag... you're it...
by korrelan (AI News )
Today at 02:14:50 pm
In need of a psychological coach?
by ruebot (AI News )
October 13, 2018, 01:33:37 pm
Leela Chess Zero
by Art (AI News )
October 08, 2018, 03:30:42 pm
Alpha GO Zero
by Art (AI News )
October 08, 2018, 02:48:32 pm
Detecting fake news at its source
by Tyler (Robotics News)
October 06, 2018, 12:00:55 pm
Phase Change Memory
by Art (AI News )
October 05, 2018, 01:21:59 pm
Model helps robots navigate more like humans do
by Tyler (Robotics News)
October 05, 2018, 12:01:40 pm
Regina Barzilay, James Collins, and Phil Sharp join leadership of new effort on machine learning in health
by Tyler (Robotics News)
October 04, 2018, 12:00:23 pm

Users Online

44 Guests, 2 Users
Users active in past 15 minutes:
LOCKSUIT, korrelan
[Trusty Member]

Most Online Today: 108. Most Online Ever: 208 (August 27, 2008, 09:36:30 am)

Articles