The stink on instincts

  • 27 Replies
  • 10609 Views
*

Freddy

  • Administrator
  • **********************
  • Colossus
  • *
  • 6860
  • Mostly Harmless
Re: The stink on instincts
« Reply #15 on: February 20, 2016, 04:04:17 pm »
That video of early attempts to fly was hilarious - the one at the end though was a bit alarming !

*

Zero

  • Eve
  • ***********
  • 1287
Re: The stink on instincts
« Reply #16 on: February 20, 2016, 05:52:24 pm »
Sure, they were brave men, risking their lives so we can fly...  :idiot2:  O0

I'd like to go further on the "get closer / get away" couple. It's really a primary function.

That's what you do when you wonder "Will this cake be ready before my wife comes home?", you open a sub-workspace on "when does she come home?", and when you answered the second question, you can answer the first one.
Or, while you're driving, someone asks you something complicated, you say "wait a minute", you focus on the difficult situation ahead of you, and when it's past, you can go back to what was asked.
It's always a "get in / get out" kind of action.
It's also how we modify settings on our smartphones: we get in the menu, maybe it's in there, get in, nope, get back, try another one, that's it, done, get back, get back.
It's also how we use a web browser. Ask big bro', get the list, maybe this one, open in new tab, nope, get back, try another one, ...etc.

So this "get closer / get away" thing seems like a fundamental way of functioning to me. Some kind of elementary action particle.

*

8pla.net

  • Trusty Member
  • ***********
  • Eve
  • *
  • 1307
  • TV News. Pub. UAL (PhD). Robitron Mod. LPC Judge.
    • 8pla.net
Re: The stink on instincts
« Reply #17 on: February 21, 2016, 06:12:30 pm »
Mission: Impossible - Cat Protocol

My Very Enormous Monster Just Stopped Using Nine

*

Freddy

  • Administrator
  • **********************
  • Colossus
  • *
  • 6860
  • Mostly Harmless
Re: The stink on instincts
« Reply #18 on: February 21, 2016, 07:41:55 pm »
 ;D

Seems like his instinct to survive was a bit lacking there.

*

Zero

  • Eve
  • ***********
  • 1287
Re: The stink on instincts
« Reply #19 on: February 22, 2016, 02:58:51 pm »
 :2funny: get closer, nope, get baaaaaaaaaaaaaaaaaaaaaaaaaaaaack

*

DemonRaven

  • Trusty Member
  • ********
  • Replicant
  • *
  • 630
  • Disclaimer old brain @ work not liable for content
    • Chatbotfriends
Re: The stink on instincts
« Reply #20 on: February 23, 2016, 06:33:10 am »
An instinct is something you don't need to learn — it happens naturally, without you even thinking about it. Not everything cats do is actually instinct. I raised some kittens twice. First time from birth, the second time from the age of 3 weeks old. It was fascinating to learn what they picked up from me and the other pets. They imitated a clucking noise i made to them when they were kittens and even shook their heads at their kittens like I did when they got old enough to have babies. One of my cats that was raised with a couple of dogs actually learned to do doggy tricks like sit, beg and roll over.

In any case instinct is overrated in my humble  opinion. There is machine learning. The trick is getting it to learn like we do without intervention from humans.
So sue me

*

djchapm

  • Trusty Member
  • **
  • Bumblebee
  • *
  • 37
Re: The stink on instincts
« Reply #21 on: February 25, 2016, 03:45:08 am »
Hey just got back and caught up.  Really enjoy the discussion - and I'm very impressed with Korrelan's work.  And his way of thinking - I believe I'm more in that camp that true learning, something from nothing, and not a 'directed' approach is how true AI will emerge. 

Really fascinating post Korrelan - and loved the videos as well, I'm guessing as much time was spent on visualization as was spent on the wiring/building of the network!  Impressive.  Is this all yours or what is the project?

I think you've tackled some major issues - like constant streaming data and mixed signals, 'frame of thought'.  Do you feel that is enough to identify situations and multiple events happening at the same time with comprehension?  Seems there would be a challenge in seeing the difference in similar situations.  Or you might get into a situation where you have never ending growth of the network trying to maintain separation.

I may not be making a lot of sense - happens when I start thinking.  You have any info on where we could read into your project more?

*

Korrelan

  • Trusty Member
  • ***********
  • Eve
  • *
  • 1454
  • Look into my eyes! WOAH!
    • YouTube
Re: The stink on instincts
« Reply #22 on: February 26, 2016, 10:29:33 am »
Quote
I believe I'm more in that camp that true learning, something from nothing, and not a 'directed' approach is how true AI will emerge. 

Good to see someone who agrees lol.

If we can build a machine that can learn anything and experience reality as we do… it should be able to learn to be ‘Human’.

When I started the only human cortex area that had been really studied, and the research peer reproduced, was the visual areas.  So I designed a connectome to reproduce the V1 map patterns found in primates.  I figured because most of the cortex has a similar repetitive structure, if I could get this working the rest would fall into place.  When I tested my visual network and it also recognized audio/ tactile sensory streams, it spurred me on; been at this for long, long time lol. 



Monkey V1 cortex map with a section of my map (top left) overlaid to show the similarity.



Vid showing real time creation of map from visual data.

Quote
I'm guessing as much time was spent on visualization as was spent on the wiring/building of the network! 

Yeah!  The visualization is important; it’s how I work best.  The vids are from the CAD software I designed to build the connectome in 3D virtual space.  I have written analytical routines to track the learning/ logic of the system but I still prefer to see what’s happening in real time.  I can freeze, rewind, etc and actually move/ view deep within the connectome and see/ edit what’s happening.



This example shows how a trained area of cortex responds to the test patterns I'm feeding in.  You can see the relevant areas around the output pyramid cells actually firing correctly.  Visualization makes the job much easier.

Quote
Is this all yours or what is the project?

It’s my own personal spare time project at the moment. 

Quote
Constant streaming data and mixed signals, 'frame of thought'.

I use loop buffers on the audio/ visual/ tactile/ joint position/ etc input streams, so the system has time to catch up if it lags behind on the processing.  Also means I can freeze the simulation at any point and debug synapse etc. I designed the program to run on multiple cores and cpu’s so I can add as many processors as I need as the complexity grows. I’ve also written auto-balancing software that shares the workload quotas across cores based on the complexity/ time cycle of cortex areas.  I’ve also just increased the overall speed by realizing that a lot of problems I’ve encountered, can be solved by the system going into a sleep cycle.  It needs frequent sleep… and it dreams.

Quote
Do you feel that is enough to identify situations and multiple events happening at the same time with comprehension?  Seems there would be a challenge in seeing the difference in similar situations.

The system grows and adapts to the complexity of its experiences.  Because each experience is broken down into its finite facets there is always a deeper resolution/ difference the system can exploit to tell the frames/ chains apart.  The seed doesn’t even know the difference between light and dark at first, it takes time to build the hierarchical networks/ knowledge/ feedback required to recognize complex concepts etc.  I use Tv, audio books and human interaction to train my AGI.

Quote
Or you might get into a situation where you have never ending growth of the network trying to maintain separation.

The opposite problem to separation applies.  The more information is soaked up by the system the finer detailed the neural maps become.  Once saturation point is reached the whole outer cortex (not the inner) has to grow in surface area to allow neurogenesis to insert more neurons at areas of high activity to increase the resolution of that area.

This is also part of the mechanism that produces prediction/etc, because the system is self organizing similar facets of ‘thought’ patterns tend to cluster together on the cortex surface.  So when a new neuron arrives at its location it waits, sends out dendrites and connects to neighboring networks, sometimes these networks will always fire in a specific sequence, this always follows that… the neuron will short cut the sequence.  This allows the system to glean a prediction before something happens as well as still process all the relevant data that produced the patterns in the first place.

All available cortex real estate is used, so if the system is blind it adapts and the visual areas are used to enhance audio/ sensory resolutions. This was a bonus I gained from my connectome rules/ design.

Quote
You have any info on where we could read into your project more?

I have been working on a site that explains my progress but it’s not finished yet.  There are only a few vids on my Youtube page that show my early work on audio visual recognition.

Thanks for your interest; it helped me clarify my own mind.
« Last Edit: February 26, 2016, 12:03:40 pm by korrelan »
It thunk... therefore it is!...    /    Project Page    /    KorrTecx Website

*

Korrelan

  • Trusty Member
  • ***********
  • Eve
  • *
  • 1454
  • Look into my eyes! WOAH!
    • YouTube
Re: The stink on instincts
« Reply #23 on: February 29, 2016, 11:09:43 am »
Hi all…

I’m not sure if this is the correct place to post this but it kinda links on from me explaining a bit about my AGI. 

Part of my research involves making my AGI experience human traits/ problems so I can see if my simulation is an accurate model of the processes running the human brain. 

Yesterday I was messing around with synesthesia.  Remember when the research came out that linked the way people saw certain letters of the alphabet in specific colours. 99% of them saw the same colours as a popular childs toy of the time, the Fisher price fridge magnets.  They had learned the alphabet at an early age from these fridge magnets and the colours just stuck with them.

http://neurocritic.blogspot.co.uk/2013/01/fisher-price-synesthesia.html

Once I had reproduced this effect I moved on to Hypnotism. I had an idea of how it works and how to reproduce the effect in my system.  Basically the higher cognitive functions run at a much higher frequency then the main deep ‘executive/ working’ networks. So there had to be a way to induce a rhythm in the system that would enable the main controlling network to run but inhibit the higher cognitive functions.

After many glasses of rum and messing about I found there was a way… if I induced a harmonic low frequency in the lower networks they would function correctly but the higher frequency networks would be out of step, and thus unable to effect the system. I had a kind of awake yet sleeping zombie system.

Now… the only way to get the system to be susceptible to a low frequency harmonic was to introduce the experience/ learning at a very early stage in its development.

I know this bit is weird but… I don’t dream very often but last night I had Derren Brown rocking a long chain of infant cots joined together with rope, all the infants where crying, then one by one they fell asleep.

Which brings me back to synesthesia and how early learning can instill traits that affect us in later adult life.

I now have the theory that people who are more susceptible to being hypnotized where regularly rocked to sleep as babies, their parents where training them to fall into an induced sleep/ altered state when they experienced a rhythm.

Any ideas/ experiences/ thoughts?  :)


« Last Edit: February 29, 2016, 12:25:36 pm by korrelan »
It thunk... therefore it is!...    /    Project Page    /    KorrTecx Website

*

ranch vermin

  • Not much time left.
  • Terminator
  • *********
  • 947
  • Its nearly time!
Re: The stink on instincts
« Reply #24 on: March 19, 2016, 10:34:09 pm »
Birds flap wings. Planes don't.
Quit the monkey experiments.

*

Korrelan

  • Trusty Member
  • ***********
  • Eve
  • *
  • 1454
  • Look into my eyes! WOAH!
    • YouTube
Re: The stink on instincts
« Reply #25 on: March 20, 2016, 02:30:59 am »

It thunk... therefore it is!...    /    Project Page    /    KorrTecx Website

*

Korrelan

  • Trusty Member
  • ***********
  • Eve
  • *
  • 1454
  • Look into my eyes! WOAH!
    • YouTube
Re: The stink on instincts
« Reply #26 on: March 31, 2016, 06:11:29 pm »
I was just reading back through some old posts.

I noticed Ranch said...

Quote
Quit the Monkey experiments

I do hope it didn't come across that I was experimenting on monkeys lol.

I took the research from the web and based my simulation upon it...

I don't dissect monkeys lol.
It thunk... therefore it is!...    /    Project Page    /    KorrTecx Website

*

Zero

  • Eve
  • ***********
  • 1287
Re: The stink on instincts
« Reply #27 on: April 01, 2016, 08:44:56 am »
Birds flap wings. Planes don't.
Quit the monkey experiments.

Ranch, do you mean "we should not try to simulate animal brains, but human minds"?

(wow Smartbird is beautiful)

 


LLaMA2 Meta's chatbot released
by spydaz (AI News )
August 24, 2024, 02:58:36 pm
ollama and llama3
by spydaz (AI News )
August 24, 2024, 02:55:13 pm
AI controlled F-16, for real!
by frankinstien (AI News )
June 15, 2024, 05:40:28 am
Open AI GPT-4o - audio, vision, text combined reasoning
by MikeB (AI News )
May 14, 2024, 05:46:48 am
OpenAI Speech-to-Speech Reasoning Demo
by MikeB (AI News )
March 31, 2024, 01:00:53 pm
Say good-bye to GPUs...
by MikeB (AI News )
March 23, 2024, 09:23:52 am
Google Bard report
by ivan.moony (AI News )
February 14, 2024, 04:42:23 pm
Elon Musk's xAI Grok Chatbot
by MikeB (AI News )
December 11, 2023, 06:26:33 am

Users Online

331 Guests, 0 Users

Most Online Today: 330. Most Online Ever: 2369 (November 21, 2020, 04:08:13 pm)

Articles