flat assembler
Message board for the users of flat assembler.

Index > Heap > how to teach a program to learn and become conscious

Author
Thread Post new topic Reply to topic
sleepsleep



Joined: 05 Oct 2006
Posts: 8900
Location: ˛                             ⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣Posts: 334455
sleepsleep
intrigued by the idea,

how to have a compiler or assembler that conscious, keep on try and assemble itself.

limit is an important word,
goal and continue improvement are necessary to fuel the loop,
and probably death is to exit loop and have us debug on the spot.

to add or take out something by itself with purposes,
how to evolve from basic knowledge and setting goals?

let say 2 pixels (Red), 2 pixels (White), randomly put in a 1000x1000 px spaces,
the goal is to let both met next to next,
Code:
R R
    

will cause another 1px R to exists in spaces randomly, maybe 200 px around this R R pair.

try to avoid different color, if different color could trap (square) that pixel, the pixel will become different color,

Code:
R R R
R W R
R R R
    

W will become R

but because we view from top of those spaces, we know exactly where both pixel location,

assume we put eye sight api into pixel, able to view 100 px left, right, top or bottom, diagonal with degree of certainty from 1 to 100%, since human eye sight is limited too.

we put move api into pixel, able to move or run, eg 5 px / second

we put energy api, like human body energy consumption, no move (rest) + energy,

R R pair will cause energy storage to increase 7% when travel in pair

if pixel energy drop to 0, pixel will remove from spaces, death.

could bring in health api, poison api (like bomb) planted by pixel

shit, just got an AI game.
or sounds like microorganism doing inside my body?
Post 20 Aug 2015, 21:40
View user's profile Send private message Reply with quote
revolution
When all else fails, read the source


Joined: 24 Aug 2004
Posts: 17278
Location: In your JS exploiting you and your system
revolution
There is already a massive amount of research in AI. It would probably be prudent to study what has already been done, rather than restart from scratch, and then you can make improvements and/or incorporate existing ideas into your approach. While it might be possible to create something better from scratch it would be a huge undertaking, and you would miss out on a lot of good stuff that has already been discovered.
Post 21 Aug 2015, 04:59
View user's profile Send private message Visit poster's website Reply with quote
sleepsleep



Joined: 05 Oct 2006
Posts: 8900
Location: ˛                             ⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣Posts: 334455
sleepsleep
thanks revolution,
i might need to trial and error to learn this stuff, my fundamentals on this subject is too weak to understand those researched results.

let see if i could build a javascript sample.
Post 21 Aug 2015, 15:33
View user's profile Send private message Reply with quote
nyrtzi



Joined: 08 Jul 2006
Posts: 192
Location: Off the scale in the third direction
nyrtzi
How to make a program which is conscious, self-aware and can learn? I'd say that you first need to define the concepts themselves on a more philosophical level before trying to come up with the algorithms and data structures to describe them.

You might end up needing to think about homoiconicity, reflection, reification, temporal data structures and both pattern recognition and backtracking search for prediction purposes. Then again there might smarter ways but I'm no expert. I'm just speaking on the basis of how I assume the human mind works.

And if you ever come close to making something which actually works please don't make the mistaken assumption that the artificial intelligence must be completely objective in its thinking. There are a whole lot of AIs in science fiction who like to think of themselves as objective and perfect without even understanding what those words mean. I mean the same way those "perfect soldiers", "superhumans" or "overmen" also turn out to be something other than what their name implies.
Post 21 Aug 2015, 21:21
View user's profile Send private message Reply with quote
TheRaven



Joined: 22 Apr 2008
Posts: 89
Location: U.S.A.
TheRaven
IBM Watson is exactly this and has been successful as hell in medical application assisting cancer treatment specialists research treatment opportunities for cancer patients. It is insane how fast, accurate and generally bad ass Watson is --Dr. Watson, hell yeah!

AI is sweet and IBM knows better than to go Skynet --that's end game sh!t there.

@sleepsleep: trial and error and general, incurable curiosity = passion.

OCR systems are slightly AI as is speech to text --edge detection for best match; comparisons are generic, atomic constituents of AI collectively known as reasoning.

It's called AI, but we really mean and hope for reasoning.

_________________
Nothing so sought and avoided more than the truth.
I'm not insane, I know the voices in my head aren't real!
Post 20 Oct 2016, 05:39
View user's profile Send private message Reply with quote
sleepsleep



Joined: 05 Oct 2006
Posts: 8900
Location: ˛                             ⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣Posts: 334455
sleepsleep
@TheRaven, passion is hard to find Wink
Post 20 Oct 2016, 07:16
View user's profile Send private message Reply with quote
YONG



Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E
YONG
sleepsleep wrote:
passion is hard to find
Give revolution some cash and he/she will get the job done with passion! Laughing

Wink
Post 20 Oct 2016, 08:20
View user's profile Send private message Visit poster's website Reply with quote
revolution
When all else fails, read the source


Joined: 24 Aug 2004
Posts: 17278
Location: In your JS exploiting you and your system
revolution
Passion comes from desire and enjoyment, not cash. Only responsibilities come from cash. However I also like to eat, so cash is a requirement, and if I can combine that with something I desire and/or enjoy then all the better. I wonder if an AI would think the same way?
Post 20 Oct 2016, 16:35
View user's profile Send private message Visit poster's website Reply with quote
YONG



Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E
YONG
revolution wrote:
Passion comes from desire and enjoyment, not cash.
Really?

In your career, how many times have you pretended to be passionate about a big project before your potential client?

You probably have lost track of the count already.

Wink
Post 21 Oct 2016, 04:02
View user's profile Send private message Visit poster's website Reply with quote
revolution
When all else fails, read the source


Joined: 24 Aug 2004
Posts: 17278
Location: In your JS exploiting you and your system
revolution
YONG wrote:
revolution wrote:
Passion comes from desire and enjoyment, not cash.
Really?
Yup.
YONG wrote:
In your career, how many times have you pretended to be passionate about a big project before your potential client?

You probably have lost track of the count already.
Maybe once or twice, when I was more naive than I am now. Ultimately people see through it, and it backfires when the drudge work begins. So not any more, life's too short to be pretending all the time.
Post 21 Oct 2016, 06:29
View user's profile Send private message Visit poster's website Reply with quote
YONG



Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E
YONG
revolution wrote:
life's too short to be pretending all the time
Good point!

So, why don't you just reveal your genuine gender to the forum members?

No more guessing. No more pretending. Let us know who you really are!

Wink
Post 21 Oct 2016, 06:46
View user's profile Send private message Visit poster's website Reply with quote
revolution
When all else fails, read the source


Joined: 24 Aug 2004
Posts: 17278
Location: In your JS exploiting you and your system
revolution
YONG wrote:
No more pretending.
Life's too long to have no pretending at all. There has to be some room for a bit of mystery and uncertainty.
Post 21 Oct 2016, 08:26
View user's profile Send private message Visit poster's website Reply with quote
YONG



Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E
YONG
revolution wrote:
Life's too long to have no pretending at all.
Cool! You're playing both sides of the fence.

client: Can your application deal with the bottom 0.135% of the population?

revolution: Nope. That is over 3 standard deviations to the left. We call them the lower-extreme cases -- they never pretend to be anything/anyone. To me, life is too long to have no pretending at all.

client: How about the top 0.135% of the population?

revolution: Nope. That is over 3 standard deviations to the right. They are the upper-extreme cases -- they always pretend to be something/someone. To me, life is too short to be pretending all the time.

client: Duly noted.

revolution: So, would you buy our application?

client: I will let you know when the time comes. To me, life needs some room for a bit of mystery and uncertainty.

revolution: What?

Wink
Post 21 Oct 2016, 11:42
View user's profile Send private message Visit poster's website Reply with quote
revolution
When all else fails, read the source


Joined: 24 Aug 2004
Posts: 17278
Location: In your JS exploiting you and your system
revolution
So, umm, does that mean life is just the right length to have some pretending some of the time? Is that something to do with balance and/or diversity? I dunno, but it sounds about right. And yeah, those outliers are always the problem ones. Just like in code with those pesky corner cases that seem to take up half the code for very little gain but still have to be there else the the whole thing fails.
Post 21 Oct 2016, 12:12
View user's profile Send private message Visit poster's website Reply with quote
Display posts from previous:
Post new topic Reply to topic

Jump to:  


< Last Thread | Next Thread >
Forum Rules:
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You can attach files in this forum
You can download files in this forum


Copyright © 1999-2020, Tomasz Grysztar.

Powered by rwasa.