• FEATURED

View more

View more

View more

### Image of the Day Submit

IOTD | Top Screenshots

### The latest, straight to your Inbox.

Subscribe to GameDev.net Direct to receive the latest updates and exclusive content.

# HOW TO BRUTE FORCE INTELLIGENCE

Old topic!

Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.

11 replies to this topic

### #1/ rouncer   Members

Posted 03 May 2014 - 11:59 PM

ok.  it might not work.
dont get your hopes too high up, or youll come crashing down.

it all revolves around an interactive video technique,  and that might not work... see i can control myself inside a video record, by biasing the video to the control, but keeping it most similar in the sensor itself,
if that doesnt work, it doesnt work.  this is how you pull your simulations with full physics, everything.

ok, then.  if that works, and you have enough samples,  he can simulate motor mutations.

ok.  then the idea is follows.

mutate as much logic as you can, (that when tested! thats the only way youll know if its a decent conjunction.), will arise at the moment of the triggering of a motivator, and nullify a demotivator.

then leave that going till it hits stasis, set that in rock, or maybe it keeps learning, and the level above adapts to changes in the level before. (just like your sensory record itself)

but the idea is, you can then mutate for logic (in this new level) that will bring up the previous levels logic, and the speculation goes, theres more of it now?
and its further back conceptually (not to do with order in time, totally, but order in time is a part of it, how it simulates the situations) from the instinctive moments. so the procrastination goes.

its a piramid/heirarchy scam, the same. hehe

Edited by rouncer, 04 May 2014 - 12:01 AM.

### #2Waterlimon  Members

Posted 04 May 2014 - 04:08 AM

POPULAR

a please into more of coherence

o3o

### #3Erik Rufelt  Members

Posted 04 May 2014 - 04:15 AM

POPULAR

hehe. yes

it might not work.

ok, but the idea is work. the previous logic will always be leveled up for the next speculation, always more

but petaflops is few in todays climate to necessary function.

please blog about development process so all can share in future the same

but i think no rock but let always learning conceptually further, but FORWARD

### #4samoth  Members

Posted 04 May 2014 - 04:39 AM

POPULAR

Markov model generated text?

### #5/ rouncer   Members

Posted 04 May 2014 - 04:49 AM

i spose, any deranged monkey could have come up with that, back to the drawing board.

Posted 04 May 2014 - 05:06 AM

POPULAR

Rouncer,

You keep sharing unusual and entirely theoretical ideas that you haven't tested (examples 1, 2 & 3), and you don't explain them very well.  Did you even notice that none of the posts before mine were actually proper feedback?

Almost a year ago I suggested that you might be better off if you actually test these ideas so that you have something to show, and that you should put some effort into better formatting your posts.

People will forgive small errors if English isn't your first language, but honestly it seems like you're barely trying (basics like no capitalised letters at the start of sentences for example), and you're constantly sharing completely untested (and often very unusual) ideas and then just belligerently insisting you must be correct if people question you.

This is simply not a constructive way of posting, and with some minimal effort you could address these issues and actually have a sensible conversation people will be interested in.  It's clear that you can sometimes produce potentially interesting results, and it would be really nice if you could have proper conversations with people and get some real input into your ideas so that you can potentially develop them further.  It would only take some minimal effort on your part to format your posts better and to perform some rudimentary tests on your ideas rather than posting while they're still purely conceptual.

Sorry if that came across as harsh, but I think it's something you need to be told, and I sincerely hope it helps you.

My suggestions, in summary:

• Take the time to try to write proper sentences, beginning with capital letters and ending with periods.
• Take the time to try to structure your posts into a sensible order rather than just rambling your thoughts and immediately hitting the submit button.  After typing your posts, take a moment to read back over them and edit anything that isn't clear before you submit them.
• If possible, actually test your ideas or explain relevant background material so that people can see that it isn't just theoretical musings that may not amount to anything.

Edited by jbadams, 04 May 2014 - 05:14 AM.

### #7/ rouncer   Members

Posted 05 May 2014 - 12:05 AM

Jbaddams, all 3 posts are the same idea, im just slowly developing the implementation,  they are non contradictory, just said at different levels of realization, maybe im wrong for telling fibs before ive prooven it works, like its no use to anyone, no matter how much i think it works.

Its just I love ai the most, if I could make my game play itself with symbolic concepts, and even generate athletic liquid motor at the same time, I could play his brain like a video game at the end and be happy with my life achievement.   im sure harder things have been done.

I thought some more about it... but anyone can think about this if they want. (i have no genius complex, i feel a little foolish actually)

And, just say youve compressed analyzed a stream into packets of common repetitions, and these in a hierarchy.

just say you set up a further grouping system, which can group all up the hierarchy in a single group if it wants, and these groups connect with

inclusive and disclusive and even xor connections, which then fire if its recorded that if that logic group comes up - its definitely going to bring up a motivator or demotivator.

then you just get relations of those relations, and relations of those relations.

a cat in a hat in cat in a hat in a cat a hat, persay.

then hopefully, your robot sees another robot get run over by a car, and it triggers a demotivator, because it connected logic gate pointer to logic gate pointer, to some bad moment it experienced,  thats the idea, after all the correlations between all sprites, has been discovered, then it knows if its to be done or avoided, without all the time having to fail at it first, exactly, if it develops a relation pointer instead.

and its these correlations which supervise the motor,  judge it good, to be an accepted change, that means appropriate correlations came up, or judge it shit, a terrible motor example... because either no good correlations were there, or a bad correlation was there.

do you think it has potential?  the robots brain is just on record.

Edited by rouncer, 05 May 2014 - 04:29 PM.

### #8DanielKruyt  Members

Posted 30 May 2014 - 02:39 PM

The writing that you find when your cat took a nap on your keyboard makes more sense than this guy.

### #9LightOfAnima  Members

Posted 01 June 2014 - 05:47 PM

Hey, have you ever looked into machine reinforcement learning? It sounds slightly what you are talking about, I think .-. It can certainly be effectively applied to 2D games from what I hear, and surely to 3D as well. And as another person mentioned, there does see to be a slight resemblance to the Markov model, which could be used to help realise your goal

### #10comfy chair  Members

Posted 09 June 2014 - 08:07 AM

Congratulations to rouncer's creators - I think you just beat the Turing test.

### #11Tutorial Doctor  Members

Posted 09 June 2014 - 09:45 AM

Haha comfy chair. That is too funny. Brute force intelligence indeed.

They call me the Tutorial Doctor.

### #12ApochPiQ  Moderators

Posted 09 June 2014 - 11:45 AM

Alright, guys, enough piling on the poor OP. This can stop now.

Wielder of the Sacred Wands

Old topic!

Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.