Ai Dreams Forum
AI Dreams => General Chat => Topic started by: ranch vermin on August 27, 2015, 05:48:42 am
-
My theories coming together, and im starting to realize the time for immature brag sharing is over... ive just privatized all my videos. Im not sure if I want to help people develop ai anymore - its for me to do.
Do any of you guys have some top theories that you never disclose - in case they fall into the wrong hands?
I really feel like I really am a NASA guy now, or someone like a nasa guy, and its kinda ordinary - making advanced technology out of your bedroom.
-
Oh now, this is a fun discussion. It wouldn't surprise me if most of our community had private A.I. projects!
-
Real brag is believing that what we do is advanced enough to help "wrong hands". Wrong hands probably don't need us to do wrong things. Now, walking the path together is probably more fun than reaching the destination alone, isn't it?
And, AGI is already here, we name it Internet.
How about simulating a cat? That would be fun :uglystupid2:
-
No! Cats HATE to be simulated!...just ask one and notice the look you'll get! :2funny:
Instead, let's simulate an honest politician! Couldn't hurt cause we know such a thing doesn't exist! O0
-
Honest politician? Now that is a real challenge! :2funny: Seriously, I think quantum physics says that such a thing cannot exist.
But think about the cat. You can simulate it with various degrees of realism. First you can make a chatbot:
You: hey kitty.
Cat: meow.
And if you're a nasa guy 8), say you can accurately simulate its whole brain. Then, in what world would this cat live?
-
Perhaps im being a little silly but I'm thinking my theory is too powerful to share... how confident are you guys with your theories?
F.E. Im pretty sure I could automate a whole factory, possibly the supply also, so its pretty hardcore stuff - but it remains a theory - I havent fully implemented it yet, (im busy at work on my computer now!) but im feeling a little too lucky for my liking.
-
How is your theory? Is it mathematically powerful? Does it involve crunching a lot of data very quickly thanks to a "magical" algorithm, or storing a lot of data in a very small place?
To answer your question, IMHO, being confident in your own theories is ok: you're supposed to be confident, it means that it's time now to switch from theory to implementation. Feeling that it's too powerful to share is ok too: it means that you care, because you're a good man. In my opinion, you should implement your work on a computer that is not connected to internet, and then decide whether you'll share it or not. I'm sure you're aware that implementing things is important for theory development. I wish you the best.
-
I was thinking leaking a little bit of it, but I just have to say - being a forum user, isnt the greatest thing to do - holding your technique safe.
Maybe I should go visit some casual forums, instead of technical ones, because its become fruitless me using them, I think it always was, I just liked bragging about my ideas, like an immature idiot.
I think maybe you guys wont see much of me anymore.
It was nice sharing my work with you so far, thanks alot guys, it was fun. :)
-
how confident are you guys with your theories?
I'm quite confident I'm working along the right lines. I've basically back engineered how I think the human brain works. Over the years I've distilled my theories down into one design of spatio temporal neural net.
My system can recognize faces and objects, parse/ understand speech in any language, control servos (motor cortex), naturally runs internal simulations before it acts to predict the consequences (sub conscious), etc...it even needs sleep... and it dreams. I think it possesses most of the traits we endow to being intelligent. It experiences the world through its own senses, microphones, cameras, torque sensors etc and I'm in the process of building a tactile/ feedback body etc.
I'm having problems getting it to concentrate on what I'm teaching it at the moment though. Its attention networks arn't fully developed yet so it does/ listens to what it finds the most salient/ interesting input. It will stare at a clock face for hours... ooo... it moves lol.
I left it listening to HG Wells 'The War of The Worlds' to help fine tune its audio cortex regarding human speech, after the session I checked the main audio cortex map and noticed most of the new synapse where in the 5 - 12k areas... not human speech.
I'd left the door undone, and we live in the country side. Apparently I now posses an AGI that's a expert on bird songs.
I must admit... at the moment I feel like I've designed a psychopath lol.
Edit: Just for fun I isolated the 'bird song' mapping sensitivity in the audio cortex, the grey bar at the bottom compared to other pattern frequencies learned.
https://www.youtube.com/watch?v=Abs_DKjiZrM (https://www.youtube.com/watch?v=Abs_DKjiZrM)
-
Ranch,
Don't leave! Your input is valued and appreciated! This is a place for like-minded people or else we wouldn't be having this section at all.
Automated warehouses are certainly not new but the degree of automation certainly might be.
Perhaps the day will come when a large factory or warehouse only has a minimal crew / staff for maintenance or breakdowns, etc.
This video was over 7 years ago. I didn't have time to research more current tech.
https://www.youtube.com/watch?v=lWsMdN7HMuA (https://www.youtube.com/watch?v=lWsMdN7HMuA)
Come on...at least give us a teaser or two....
-
Yes please stay Ranch!
-
Ranch...
You don't have to tell everyone exactly how your theory/ code works.
The forum is very useful for getting ideas/ feedback/ news etc and posting demos.
Besides, the human element of occasionally breaking from coding and actually conversing with another human helps me keep my sanity... just.
Also, we all enjoy a pat on the back every now and again, it helps drive and inspire us... at least we understand the complexities and effort provided to such a project.
I've done the locked away... don't tell anyone... thing... it ain't healthy.
Keep the secret... but show the results :)
Edit: See even my AGI knows you...
https://www.youtube.com/watch?v=65xYHWlXKM0 (https://www.youtube.com/watch?v=65xYHWlXKM0)
I tried tapping the monitor to gets its attention... it just looked at me gone out. So I disconnected its camera and manually pointed it... the machine just had a 'out of body' experience... haha... it ain't happy lol. No head positioning data to correlate with camera movements... it's jumpy because it can see other objects in its peripheral.
-
Protect your theory by thoroughly understanding it.
In theory, your theory may have a theory about you.
Your theory may be protecting its theory from you.
-
Congratulations korrelan,
That's really super cool ! Don't feel compelled to answer if this is TOP SECRET...
I strongly respect your privacy... But, is that cross hair for human or machine?
It's a computer science question. Like FORTRAN cards: Only the holes are read by
the machine. Anything printed on the cards are for humans.
I think FORTRAN cards may be applied to implementations of our theories, where
we map the part only the machine knows to the part only the humans know.
Of course, I know the answer already, in theory. By just asking the question, "Is the cross hair for the gun or the shooter?"
-
Yar TOP SECRET... lol.
The cross hair is a remnant from when I used to manually guide it, although the camera/ eye has a higher resolution of receptors toward the fovea/ center. I've noticed that the AGI doesn't always use it when paying attention/ recognizing an object.
Saccades are induced as part of the video stream pre-processing... this might be causing it. The system is kind of holographic by design, so all cortex areas receive stimulus from all sensory inputs initially, then either integrate/ use or discard the information through experience. So it could be an audio imbalance effecting the eyes or an emotional remnant, I've been messing around with flushing the lobes with various equivalent neurotransmitters which alter the functionality of the synapse based on the sensory stream (mood/ feelings)... it's just a really quirky system lol.
Edit: This is the AGI responding to emotional inflections in my voice using the audio cortex only...
https://www.youtube.com/watch?v=vt8gAuMxpds (https://www.youtube.com/watch?v=vt8gAuMxpds)
You wouldn't want it driving your car... road... road... oh! look rabbit... must hug bunny... crash... lol.
-
Friend,
Please let me share a simple suggestion.
One, that I plan to implement for my A.I.
to use in place of audio. A mute button!
A two line Closed Captioning System.
(Frame#.) (CC text)
0. (empty, not shown)
1. THIS STARTS
(1. moves up one row)
0. THIS STARTS
1. THIS IS NEXT
(1. moves up one row)
0. THIS IS NEXT
1. THIS IS FINAL
(1. moves up one row)
0. THIS IS FINAL
1. (empty, not shown)
A word wrap function is needed.
Text should have a foreground
and a background color.
-
This is the AGI responding to emotional inflections in my voice...
youtube.com/watch?v=vt8gAuMxpds (https://www.youtube.com/watch?v=vt8gAuMxpds&rel=0&controls=0&autoplay=1)
This golden ratio is one of my all time A.I. favorites!
-
Korrelan, I have no idea what it's doing but it looks cool ;D
-
Cheers chaps :)
The top right is a spectrograph of the vocal input. I take the microphone input and run it through a fourier series analasis which breaks it down into the individual frequencies over time. I also needed the power/ volume of the input, shown very top right.
Bottom right... The output from the spectrograph is fed into my neural model, this small section used for the test comprised of 6 sheets of 300 * 200 neurons so... 360K neurons ish. Each neuron is synapsed to every other so... 129600000000 ish synapse.
I needed a simple face for expressions, so I adapted the 'golden ratio' face to a vector set. Then re-mapped to produce simple expressions and wrote the animation routines to smoothly move from one to another.
I trained the neural net on normal speech patterns using audio books, I then spoke to the system using various inflections in my voice, anger, joy and noted which areas of the neural net responded to the different inflections. I then simply used a blackboard technique to map the neurons for each inflection to the appropriate animation.
The facial expressions will eventually run through the motor cortex to express how the system is feeling, this was just test to see if I could pick up a jist from the users voice.
I did the same thing with facial ques and a combination of both... kinda freaky when your talking to the AGI and it mirrors you...
-
Impressive, indeed. Is there a website or something where I can read about your system?
-
Hi Zero
Yeah! Millions of them... neuroscience and biology sites lol... Soz couldn't resist.
Only my YouTube page, some vids of the initial neural seed I 'grow' the systems from, audio visual maps from early development runs etc... I've been working on AI for over 30 years...
Here's a vid showing my initial tests on neuro-genesis and migration, you'll see neurons moving/ zipping through the cortex layer to find their ideal position/ location amongst other tuned neurons... to enhance resolution required in that area (part of the memory/ learning)... visual orientation map I believe focused on a pinwheel..
https://www.youtube.com/watch?v=lF4WhCINHKE (https://www.youtube.com/watch?v=lF4WhCINHKE)
I'm a dreamer too... I'd like to think that one day my AGI's will explore other worlds, and as they scream through the atmosphere heading downwards toward a fresh new world to explore... if you listened very... very... carefully you would hear...
Weeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee!... lol
My take on AGI is.. just design a machine that is capable of learning anything... and then teach it to be human. :)
-
You should train your baby so it can drive a DeltaGlider in Orbiter! :jump:
Did I understand correctly? Are you handling 129.600.000.000 synapses real-time?
-
The top right is a spectrograph of the vocal input. I take the microphone input and run it through a Fourier series analysis which breaks it down into the individual frequencies over time. I also needed the power / volume of the input, shown very top right.
Bottom right... The output from the spectrograph is fed into my neural model, this small section used for the test comprised of 6 sheets of 300 * 200 neurons so... 360K neurons ish. Each neuron is synapsed to every other so... 129600000000 ish synapse.
My theory, and it is just a theory, is that both these cases are doing the same thing. If these are two ways to fill a curve (in theory), then that leaves us with the training. This is so educational. Thanks for your expert response.
-
Ok! It's Sat night here and I've been on the Rum & Malt so ignor any bad English or spelling mistakes... my brain works like... alcohol at night to brain storm... coffee in the morning to focus... and all that.
@Zero
Replace them full stops/ decimal points with commas and...
Nah! lol... miss calculation... 300 x 200 all synapsed = 60k x 60k x 6 layers = 21600000000 and... inter layer axons = 60k x 6 = 360k so... 21600360000 possibe synapse.
But... that's just a small section of cortex so way... way... frikken waaaay more... and yes in real time, though that's only possible synapses (but all are possible). It has to be an all to all connectome to start with, you only cull whats not used over time...
@8pla.net
Getting the frequencies laid out nice and neat is one thing... matching the trillions of possible combinations in real time is another, but yes... they are just simple patterns in a time frame.
Learning is a very simple concept... a system has to be told, or experience an event, it has to record the event in its 'own' terms so it can recall it, as 'it' (not you) experinced it, through 'its' own senses.
Then it just needs guidance... do I keep this knowledge and build upon it, or discard it.
I'm gonna read this in the morning and think... WTF was I rambling about lol. And one more... (just for good measure) the (...) is a really frikken weird habit I've acquired... meh!
Edit: My way is not the only way... It's just what I deem and have found to be correct... there is more than one way to fly.
-
@8pla.net
Getting the frequencies laid out nice and neat is one thing... matching the trillions of possible combinations in real time is another, but yes... they are just simple patterns in a time frame.
Learning is a very simple concept... a system has to be told, or experience an event, it has to record the event in its 'own' terms so it can recall it, as 'it' (not you) experinced it, through 'its' own senses.
Then it just needs guidance... do I keep this knowledge and build upon it, or discard it.
You are entitled to a few. Rum beverages and mistakes, that is. They go good together. I made a nice mistake myself. Instead of "fill a curve" I meant to say "fit a curve", I think. Oh, now I'm confused. No worries, when morning comes, seems to me, your response will be just fine.
-
I'll still be here. Maybe ill have to admit my ai idea isnt that original or hard to come up with. Just going through a phase. :)
-
I bet it's hard to implement though. I don't place much value in theories unless I've at least made a working prototype of it. No matter how good the idea, someone still has to do years of work to make it happen, and quite often some overlooked details pop up to make it that much harder.
-
definitely true my friend.
-
21600360000 possibe synapse.
But... that's just a small section of cortex so way... way... frikken waaaay more... and yes in real time, though that's only possible synapses (but all are possible). It has to be an all to all connectome to start with, you only cull whats not used over time...
Yeah ok, but that's still 21 billion real time, you're fkin fast!!! How do you do that? (some "don't think you are, know you are" trick?)
I'll still be here. Maybe ill have to admit my ai idea isnt that original or hard to come up with. Just going through a phase.
It's probably a very good idea. Trust mankind, share it!
I'm so happy you're staying here. O0
-
Hey Korrelan, I just had a peak at your videos. Did you craft that animating vector portrait yourself? Its pretty nice.
[EDIT] hahahha i saw you put me into a class on one of the videos hehe [/EDIT]
-
Morning all, I just crawled out my pit...
@Zero
Yes it is fast...
I understand your query but don't forget I have a Beowulf cluster...
Slice of cortex running in real time... all to all... I don't usually validate my work but just for you...
https://www.youtube.com/watch?v=PEem_1HHdU4 (https://www.youtube.com/watch?v=PEem_1HHdU4)
@Ranch
The design of the face is a well known face overlay called the 'golden ratio', but I mapped the vectors and animations. And yes my AGI now recognizes your logo... and speech patterns.
Edit: An isolated neural column in the connectome model, at this scale the full model would be miles wide lol. I'm able to actually fly through the cortex and edit areas... In my second 'lab' I have a 3 meter screen with AI driven gesture control and voice command interface... I'm a geek lol.
https://www.youtube.com/watch?v=MxNIrA1ovns (https://www.youtube.com/watch?v=MxNIrA1ovns)
I'll take some pictures.
-
Fair enough ;)
-
Hmmm... I've never really looked at my system by the number of calculations per parse before. I've read the above and feel I need to explain my 21 billion calculations.
For comparison... I have three HD monitors on my desk...
1920 x 1080 x 3 (rgb) = 6,220,800 values
X 3 monitors = 18,662,400 values
A decent 3D game will use tripple buffering and a Zbuffer... so...
55,987,200 + 18,662,400 = 74,649,699 values
X 60 fps = 4,478,981,940 values per second.
So even without alpha channels etc a average graphics card is shifting a minimum of 4 billion and that's just to run a game.
The term 'real time' is also ambiguous... to me it means the system can parse language as fast as a human can speak with no lag. I use a ring buffers for the audio stream so fast/ slow parsed sections even out 'over time'.
The reason I quoted 21 billion is because each neuron has the potential to effect all other neurons on the same layer during a parse in this section of cortex.
So straight away 300 x 200 neurons = 60k and each one can effect all others, which is 60k x 60k = 3.6 billion on one layer... x 6 layers etc.. nope sounds right unless my Math is faulty.
I suppose it also depends on the understanding/ definition of neurons and synapses and how you use them. :)
-
Hey Korrelan I just realized something... and the validity of your work, as I see it. we are all making the same thing!
Implant pattern, recall pattern, class pattern. <-do those 3 things and you get halfway to C3PO, minus a few extra details. :) just i didnt realize it so strongly till now!
That isnt a secret I guess, not inbetween us.
On the keeping of secrets...
I guess if it was me vs you, its who scaled it the biggest. and who got the guts to put it into custom wires and switches - hardware solution. ;)
(https://aidreams.co.uk/forum/proxy.php?request=http%3A%2F%2Fs24.postimg.org%2F4v0z9gy51%2Fgameson.png&hash=bce0111a9eedac7f9ed56308940cd123a8ad97bd)
-
Since weve seen some of your work- ill repost mine, so you can get a flavour for what your in for.
(pattern tracking)
https://www.youtube.com/watch?v=6pTf_lYeoqI (https://www.youtube.com/watch?v=6pTf_lYeoqI)
(pattern store-recall)
https://www.youtube.com/watch?v=m6qsw3pruQ0 (https://www.youtube.com/watch?v=m6qsw3pruQ0)
-
Haha... this is why I got out of teaching and corporate programming, always with the pressure for results lol.
I'm really liking the low poly vid. GPU obviously. Thats very very similar to how a human visual cortex sees the world except your matching the polys to the moving image, the visual cortex detects lines, gradients at set locations... makes for much easier pattern matching. Like looking through a sheet of fractured glass, vertices don't appear on the fixed map until the underlying video has the correct linear qualities to trigger it. This gives the brain a map of fixed line/ shape positions that don't move in relation to it's view point.
Implant pattern, recall pattern, class pattern. <-do those 3 things and you get halfway to C3PO, minus a few extra details
Yes... along with if this pattern is present and I get this pattern then generate another pattern from experience (or being told) which is the sum/ result, which will in turn cycle back into the system, etc, etc
Glad to see you stayed btw :D
Edit: For the visual cortex kinda like this, top left original, top right find lines etc, bottom left the points are set positions, the small lines rotate until they hit a relevant feature in the image (like a line)... then all you have to do is match a grid of angles to a data set to find the object.
(https://aidreams.co.uk/forum/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FYjDoUHh.jpg&hash=8d36f0611a3278ead911a4becffd21cd372fb035)
.Well... similar except the lines don't move in the visual cortex, and there more numerous, and... so no, not similar at all lol. :) Soz I can't help it... old habits die hard lol
-
You understand what im doing pretty much - We are working on similar lines. These 2 videos are incomplete, im trying to pull it all together right now - into a single USM Universal Sensor Model, then all the toys get to come out running on the same general algorythm. Thats the idea anyway.
Implant pattern, recall pattern, class pattern. <-do those 3 things and you get halfway to C3PO, minus a few extra details
Yes... along with if this pattern is present and I get this pattern then generate another pattern from experience (or being told) which is the sum/ result, which will in turn cycle back into the system, etc, etc
Ah yes, I forgot - in my first motor system (I havent even got up to it yet!!! and its been too long!!) this will be an unlearning method that will develop results/new attempts off old results/attempts.
-
Keep at it mate, don't give up no matter what. I've been at this for 30 years, like a Jack Russell with a bone lol. I WILL finish this.
When my time comes they better bury me with a terminal or I'm gonna be bored as sh*t lol.
Though I'd have transferred my consciousness by then... hahahahaha.. ha... Just call me... Victor... Victor Frankenstein... Hahahahahaaaa.... cough! splutter! Hahahaha... (yes I'm on the Rum again)
-
Yep...a small laptop on your chest, a small array of solar cells atop your headstone to keep your laptop charged. Lastly, a series of LED's on the stone to let us topside folks know..."You've got it!!"
Or...something like that.... :2funny:
-
Very interesting conversations guys.
May I ask you both, ranch vermin and korrelan, what is your goal by creating your "recognition algorithms"? I mean, you both are creating like, 1 part of a "car". Because you both only show videos of smart recognizers and similar, but not the other required parts of the "car". Do you plan to create a technology and sell it instead of creating an intellect? Or are you trying to create a human intelligence?
Korrelan says above a video of moving a camera around by hand at his monitor at avatars:
"Edit: See even my AGI knows you..."
Hahaha his AI must know me too!
-
@Ranch Vermin was kick off this forum. Also, he has suffered a set back because. He lost his
computers and his back up computers last month. And also, I think he has moved on to other
thing because he does not talk about AI, that much.
-
Much as I liked Ranch we had to do something about his inappropriate postings. The ban was lifted by me however, so if he wanted to come back...
-
Oh I see Ranch was banned. And he actually lost his work!! How do you know he lost even the backups!!?? Do you know what he lost?
-
Ranch's new handle at this time is silly bugger, 6/23/2017.
The video were he said that his stuff got stolen has been deleted.
In the comment section is the conversation following that video:
https://www.youtube.com/watch?v=uS5KxJpC6B4&t=2s
-
Ranch was a bright chap but had his problems… I think he was working along similar lines to me… shame.
what is your goal by creating your "recognition algorithms"?
Think of all the wonderful things modern computers can do, simulate worlds/ characters in games, weather predictions, spacecraft/ aircraft autopilots, chatbots, deep learning… the list is endless.
Modern computers are programmed usually in a high level language like Java/ python, which are written in C, which is written in machine code, which directly manipulates the stacks/ registers of the CPU… which runs on binary logic/ gates. So all the wonderful software the world is currently enjoying is basically 0’s & 1’s.
The modern computer is an amazing achievement and its provenance is comprised of generations of dedicated human programmers who have written the hierarchical sub routines/ functions/ libraries that each succeeding generation has built upon.
When you consider that 99% of what a modern computer has to do to achieve these feats is ‘pattern recognition’ you might start to grasp what my project is about.
Transmission protocols like TCP/ UDP etc simply send 1’s & 0’s. When you access a web page your computer has to firstly recognise the TCP stream, then the HTML syntax, then render the page, etc, this all requires pattern recognition.
If you draw a circle on your monitor using an art package does the ‘circle’ exist inside your computer? No…only a logical representation exists… there is no actual circle.
When you look at a circle… does a circle exist in your brain?
What I’ve designed and built is a neuromorphic processor. It uses neural processing instead of 0/ 1’s but still works at the lowest level of abstraction. So rather than binary I’m using neurons, synapse, compounds, etc. this bypasses the problems of data representation/ encoding, heuristics, symbolic processing, etc.
It is capable of achieving everything modern computers can do, and a whole lot more… but needs no programming… it learns all by it’s self from experience, and is not limited by human programming/ intelligence.
The same connectome can recognise phonemes, faces, encode short/ long/ episodic memories, perform prediction and regression of experiences, focus attention/ saliency, even dream (its version), etc…
I use voice/ video/ face recognition to test my designs because it’s a nice simple method to tracking my projects progress; this is why I post this type of video. You might not believe it but this is quite a complex project lol.
Or are you trying to create a human intelligence?
So no… I’m not creating just one part of a car… it’s a full intellect… and it certainly isn’t human by any stretch atm… which is good… I’m not after human level intellect… think human intelligence X billion.
Hahaha his AI must know me too!
That AGI did recognise your logo, it also understood the moment in time and the actions I took to show it the logo… it knew where it was in the lab and my proximity too it. It understood the time of day and was probably listening/ learning the music I was playing in the background.
Its stored/ saved now on my ‘RIP’ drive where I keep all the partly developed AGI’s I test… once they have experienced reality/ time/ experiences, even though it’s through their machine senses… it doesn’t seem right to just delete 'them'.
:)
-
Korrelan, I understand LOCKSUIT's question. I think what he means is:
Recognition is not the only job of a full intellect. There should be action too, goals, beliefs, knowledge, memories, emotions maybe... Do you have a particular approach on these subjects?
Is there a thread dedicated to your work korrelan, where we could explore your work? Do you want to share tech, like the code snippet of a neuron class, or do you prefer not telling too much?
-
Recognition is not the only job of a full intellect. There should be action too, goals, beliefs, knowledge, memories, emotions maybe... Do you have a particular approach on these subjects?
Oh I see…
To arrive at my design I have basically back engineered the human cerebral cortex and other brain areas. It’s my evaluation of how I think the human brain works.
To be sure I’m correct I have to simulate and test each facet of the human condition. So far I covered the simple things like audio/ visual/ attention/ saliency/ memories/ storing episodic memories and prediction etc.
I’ve wrote a simulation of a biological system. The neuromorphic processor (NP) grows inside the simulation. The logic and rules of the simulation govern how the NP grows and develops as it matures and experiences the world. The AGI is a ‘biological program’ that runs on the NP inside the simulation.
As you can imagine because the simulation governs the structure of the NP, and the structure of the NP governs the ‘thought’ processes of the AGI it’s been quite a long complex project. If I change the rules of the simulation to slightly adjust a growth trait of the NP structure, like say neuron spacing in layer three of the neocortex, the NP has to grow from scratch because the resulting NP connectome will be totally different, and as the structure of the NP comprises of the memories/ knowledge of the AGI it wouldn't be able to access those memories. Make sense?
As I perfect the NP and the ‘software’ that runs on it the system will (hopefully) develop the more ‘complex’ actions and traits like goal setting and beliefs. A lot of the ‘high level’ traits like ‘goal setting’ require that a large knowledge base comprising of memories and correct predictions be in place.
I understand what the system needs to do to achieve the more complex phenomena; it’s just a matter of time before it gets to this stage…work in progress.
Is there a thread dedicated to your work korrelan
Just my project thread and Google+
http://aidreams.co.uk/forum/index.php?topic=10804.0
https://plus.google.com/u/0/101478978164374900308
Do you want to share tech
Not at this stage in the project; I’ve put a lot into this project and it will probably be my last attempt at creating a true AGI.
I don’t mind discussing brain theory etc. & I'm pleased you enquired about my methods though.
I will keep my project thread upto date with milestones as I achieve them.
:)