Releasing full AGI/evolution research

  • 90 Replies
  • 3758 Views
*

LOCKSUIT

  • Emerged from nothing
  • Trusty Member
  • ******************
  • Hal 4000
  • *
  • 4108
  • First it wiggles, then it is rewarded.
    • Main Project Thread
Re: Releasing full AGI/evolution research
« Reply #30 on: March 26, 2020, 01:24:33 PM »
See last post.

Oh and see, told yous, regenerating/repairing is used in all AI, here it comes up again:
https://www.youtube.com/watch?v=bXzauli1TyU
try it:
https://distill.pub/2020/growing-ca/
Emergent

*

LOCKSUIT

  • Emerged from nothing
  • Trusty Member
  • ******************
  • Hal 4000
  • *
  • 4108
  • First it wiggles, then it is rewarded.
    • Main Project Thread
Re: Releasing full AGI/evolution research
« Reply #31 on: March 26, 2020, 01:40:17 PM »
they even mention "Embryogenetic Modeling"
Emergent

*

LOCKSUIT

  • Emerged from nothing
  • Trusty Member
  • ******************
  • Hal 4000
  • *
  • 4108
  • First it wiggles, then it is rewarded.
    • Main Project Thread
Re: Releasing full AGI/evolution research
« Reply #32 on: March 26, 2020, 02:09:40 PM »
i found that link btw AFTER i wrote all meh text. See I'm spot on in every way lol....translation, context, etc
Emergent

*

LOCKSUIT

  • Emerged from nothing
  • Trusty Member
  • ******************
  • Hal 4000
  • *
  • 4108
  • First it wiggles, then it is rewarded.
    • Main Project Thread
Re: Releasing full AGI/evolution research
« Reply #33 on: March 26, 2020, 03:10:04 PM »
Similarly we inhibit questions when see the outcome we wanted, which is like cells build/branching during organ creation/regeneration.



notice the link says

"The biggest puzzle in this field is the question of how the cell collective knows what to build and when to stop."

Based on context and hardcoded desires, it grows its way forward.

"While we know of many genes that are required for the process of regeneration, we still do not know the algorithm that is sufficient for cells to know how to build or remodel complex organs to a very specific anatomical end-goal"

“build an eye here”

"Imagine if we could design systems of the same plasticity and robustness as biological life: structures and machines that could grow and repair themselves."

"We will focus on Cellular Automata models as a roadmap for the effort of identifying cell-level rules which give rise to complex, regenerative behavior of the collective. CAs typically consist of a grid of cells being iteratively updated, with the same set of rules being applied to each cell at every step. The new state of a cell depends only on the states of the few cells in its immediate neighborhood. Despite their apparent simplicity, CAs often demonstrate rich, interesting behaviours, and have a long history of being applied to modeling biological phenomena."

"Typical cellular automata update all cells simultaneously. This implies the existence of a global clock, synchronizing all cells. Relying on global synchronisation is not something one expects from a self-organising system. We relax this requirement by assuming that each cell performs an update independently, waiting for a random time interval between updates"

Both local, and global shape of the context (what, and where (position)) affect the prediction.

"We can see that different training runs can lead to models with drastically different long term behaviours. Some tend to die out, some don’t seem to know how to stop growing, but some happen to be almost stable! How can we steer the training towards producing persistent patterns all the time?"

Sounds like GPT-2. When to finish a discovery sentence. Keep on topic until reach goal.

"we wanted the system to evolve from the seed pattern to the target pattern - a trajectory which we achieved in Experiment 1. Now, we want to avoid the instability we observed - which in our dynamical system metaphor consists of making the target pattern an attractor."

"Intuitively we claim that with longer time intervals and several applications of loss, the model is more likely to create an attractor for the target shape, as we iteratively mold the dynamics to return to the target pattern from wherever the system has decided to venture. However, longer time periods substantially increase the training time and more importantly, the memory requirements, given that the entire episode’s intermediate activations must be stored in memory for a backwards-pass to occur."

That sounds like Hutter Prize compressor improvement. Takes more RAM, takes longer, for better regeneration to target from Nothing (seed, compressed state).

"it’s been found that the target morphology is not hard coded by the DNA, but is maintained by a physiological circuit that stores a setpoint for this anatomical homeostasis"

We want to regenerate shape (sort the words/articles), and grow the organism/sentence as well. But avoid non-stop growth past the matured rest state goal and stop de-generation.
Emergent

*

LOCKSUIT

  • Emerged from nothing
  • Trusty Member
  • ******************
  • Hal 4000
  • *
  • 4108
  • First it wiggles, then it is rewarded.
    • Main Project Thread
Re: Releasing full AGI/evolution research
« Reply #34 on: March 27, 2020, 08:49:13 PM »
A good AGI predictor can answer this. Can you? Try!

"Those witches who were spotted on the house left in a hurry to see the monk in the cave near the canyon and there was the pot of gold they left and when they returned back they knew where to go if they wanted it back. They knew the keeper now owned it and if they waited too long then he would forever own it for now on."
Who owns what?
Possible Answers: Witches own monk/witches own canyon/monk owns house/monk owns cave/monk owns gold/cave owns pot/there was pot/he owns it
Emergent

*

krayvonk

  • Electric Dreamer
  • ****
  • 125
Re: Releasing full AGI/evolution research
« Reply #35 on: March 27, 2020, 09:05:41 PM »
I think that sentence is confusing and if the a.i. couldnt answer it probably wouldnt be so bad a thing.

The keeper of what?  u should dictate it.

Uve got something like word2vec in yours dont you - did u see my post?

*

LOCKSUIT

  • Emerged from nothing
  • Trusty Member
  • ******************
  • Hal 4000
  • *
  • 4108
  • First it wiggles, then it is rewarded.
    • Main Project Thread
Re: Releasing full AGI/evolution research
« Reply #36 on: March 27, 2020, 09:16:51 PM »
omg! someone answer it lol....it's EASY.........easy!!!!.........plz?

no w2v yet, but i'll get my own in there and i'm coding it! :)))))))
Emergent

*

krayvonk

  • Electric Dreamer
  • ****
  • 125
Re: Releasing full AGI/evolution research
« Reply #37 on: March 27, 2020, 09:31:52 PM »
If ur understanding the text so literally,  the robot would be really easy to trick.
Yeh, its fine if your catering the sentences for it, and its still pretty cool i guess if it could parse its way through that mumbo jumbo you wrote.

Giving the robot a bullshit detector is very important,   its the big "assimilation of lies" problem.   There is a solution tho... thats what more important to me right now.

But I guess all NLP has that problem just about,  so no biggie.

*

LOCKSUIT

  • Emerged from nothing
  • Trusty Member
  • ******************
  • Hal 4000
  • *
  • 4108
  • First it wiggles, then it is rewarded.
    • Main Project Thread
Re: Releasing full AGI/evolution research
« Reply #38 on: March 27, 2020, 09:40:19 PM »
it DOES flexible to the text.....to get the best answer probable to the answer can get.....u just cant answer my test above :)
Emergent

*

krayvonk

  • Electric Dreamer
  • ****
  • 125
Re: Releasing full AGI/evolution research
« Reply #39 on: March 27, 2020, 10:01:39 PM »
Monk owned the gold?

*

LOCKSUIT

  • Emerged from nothing
  • Trusty Member
  • ******************
  • Hal 4000
  • *
  • 4108
  • First it wiggles, then it is rewarded.
    • Main Project Thread
Re: Releasing full AGI/evolution research
« Reply #40 on: March 27, 2020, 10:04:56 PM »
Huuuuuuu! U got it!

GOOD JOB! AGI pet deserves a kibble treat. Pats heads lightly. Perr. Open mouth now.

Now we need the AGI TO DO THAT!!! Come on you guys! This is our last stand.
Emergent

*

krayvonk

  • Electric Dreamer
  • ****
  • 125
Re: Releasing full AGI/evolution research
« Reply #41 on: March 27, 2020, 10:08:17 PM »
What I want to know,  if its this easy to make A.G.I  how come it wasnt done years ago.  and what about this a.i. winter - lies?  To me it seems.

*

LOCKSUIT

  • Emerged from nothing
  • Trusty Member
  • ******************
  • Hal 4000
  • *
  • 4108
  • First it wiggles, then it is rewarded.
    • Main Project Thread
Re: Releasing full AGI/evolution research
« Reply #42 on: March 27, 2020, 10:11:38 PM »
Cus you gotta grab it by the balls and say "I'm gonna do it!".

I don't hear anyone hear saying it with enough courage!

That was a real typo btw!
« Last Edit: April 08, 2020, 12:50:25 AM by LOCKSUIT »
Emergent

*

krayvonk

  • Electric Dreamer
  • ****
  • 125
Re: Releasing full AGI/evolution research
« Reply #43 on: March 27, 2020, 10:42:49 PM »
Yeh well what were ppl doing in the 80's  *scratching* their balls?!?!?

The whole vietnam war went thru killed everyone, and why didnt they have drones back then - makes NO sense.

*

LOCKSUIT

  • Emerged from nothing
  • Trusty Member
  • ******************
  • Hal 4000
  • *
  • 4108
  • First it wiggles, then it is rewarded.
    • Main Project Thread
Re: Releasing full AGI/evolution research
« Reply #44 on: April 08, 2020, 12:48:03 AM »
Way I see motivation/ life is this. Young humans and animals are hyped to enjoy life, not bored of anything yet, and we should stay that way. Try to imagine a really incredible new gift just appeared in your back bedroom, and you feel really excited and want to go rush to go see it. That's true motivation. Dead motivation is giving up or feeling like you've finished the video game. Just start a new game! A reason to live is to have motivation to do so, because staying alive / motivation to do so comes 1st, not 2nd. If we were this motivated, maybe more people would understand we can get things we can't even get right now. And a longer life. Maybe then we could work hard on AGI like we really mean it.
Emergent

 


Users Online

64 Guests, 1 User
Users active in past 15 minutes:
infurl
[Trusty Member]

Most Online Today: 75. Most Online Ever: 340 (March 26, 2019, 09:47:57 PM)

Articles