Re: [agi] Re: A mathematics of concpetual relations?

Announcement: Face Editor Download: https://www.youtube.com/watch?v=p7k86wwUoJg -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T395236743964cb4b-M510dc85278656d48e8f0c6b4 Delivery options:

Re: [agi] Re: A mathematics of concpetual relations?

Steve, I finally found what you meant by "dimensionality" and I see that your thought on that might be stretched to my thinking about categorical abstractions or "types". You said, "Genuine computation involves manipulating numerically expressible value (e.g. 0.62), dimensionality (e.g.

Re: [agi] Re: A mathematics of concpetual relations?

Delta encoder: An effective sample synthesis method for few-shot object recognition: https://www.youtube.com/watch?time_continue=2=9q_toibmeo4  Nice AGI: https://www.academia.edu/37275998/A_Nice_Artificial_General_Intelligence_How_To_Make_A_Nice_Artificial_General_Intelligence

Re: [agi] Re: A mathematics of concpetual relations?

Rapid approximations of antilogs might be used with a trial and error method to make calculated guesses of a function that combines logs and original scalar (integration?) values. But the thing I am thinking about is that it might represent another class of compression methods that can be crudely

Re: [agi] Re: A mathematics of concpetual relations?

I think Newton's method of finding a root could provide another class of compression systems. With a variance of the function itself, I think the range of possible systems could be expanded in interesting ways. I don't know if that could be useful. I am really looking for something that can be

Re: [agi] Re: A mathematics of concpetual relations?

Brett, Steve has been talking about something similar. I understand the value of being able to add and subtract rates or ratios as a substitute for multiplication and division but I am wondering if this might be used to alleviate fundamental problems in comparing discrete states. I also might be

Re: [agi] Re: A mathematics of concpetual relations?

In some areas of electronics there are times when you need to have a positive voltage (or perhaps capacitance) in order to validate that a signal is actually being transmitted. So in this case, where an (actual) off state cannot be used (because it could be unreliable), it would be possible to get

Re: [agi] Re: A mathematics of concpetual relations?

What you are discussing is neural coding mechanisms. As you are aware spiking approaches use spike timing and spiking rates as one idea. I have another idea. A neuron fires as a result of the sum of the number of exciting synaptic connections minus the number of inhibitor connections exceeding

Re: [agi] Re: A mathematics of concpetual relations?

That idea did not work. You can send 1 of 6 states in 2 clock cycles with the method I was talking about but if you have a no-voltage state then you can send trinary digits and you can represent 1 of 9 states in 2 clock cycles. Jim Bromer On Sun, Jun 23, 2019 at 1:51 PM Jim Bromer wrote: > I

Re: [agi] Re: A mathematics of concpetual relations?

When a neuron fires, the axon inside voltage goes from -70 mV to +30 mV for 1 ms. The relevant signal in most neural models is the firing rate, not the individual spikes. The firing rate varies from 0 to 300 spikes per second but the information rate is only a few bits per second. On Sun, Jun

Re: [agi] Re: A mathematics of concpetual relations?

I think the effective voltage compression in the voltage/timing binary transmission model would approach 1/3 or 1/4. I cannot remember which one offhand. Jim Bromer On Sun, Jun 23, 2019 at 1:38 PM Jim Bromer wrote: > I should have said: The method that neurons form generative 'connections' >

Re: [agi] Re: A mathematics of concpetual relations?

I should have said: The method that neurons form generative 'connections' is irrelevant to the capability of neural activity to transmit data. The brain must be able to form reactions and to make choices like inhibiting or activating different kinds of reactions to some event. This means that the

Re: [agi] Re: A mathematics of concpetual relations?

iably tell? From: Matt Mahoney Sent: Friday, 21 June 2019 01:32 To: AGI Subject: Re: [agi] Re: A mathematics of concpetual relations? I disagree. By what mechanism would neurons representing feet and meters connect, but not kilograms and liters? Neurons form connections by H

Re: [agi] Re: A mathematics of concpetual relations?

Yeh Hebbian learning might be a means towards an unknown end,  and if its us, its going to be pretty amazing. -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T395236743964cb4b-M72c00068ad5e9b9ec0a103fe Delivery

Re: [agi] Re: A mathematics of concpetual relations?

I guess I should have not said that I totally agree with Steve's comment. When he said dimensions I was thinking more of types such as abstract types or something that is effectively similar. Suppose there was a non-standard, innovative mathematics that was able to effectively deal with data of

Re: [agi] Re: A mathematics of concpetual relations?

I disagree. By what mechanism would neurons representing feet and meters connect, but not kilograms and liters? Neurons form connections by Hebb's rule. Neurons representing words form connections when they appear close together or in the same context. On Thu, Jun 20, 2019, 4:14 PM Jim Bromer

Re: [agi] Re: A mathematics of concpetual relations?

The learning method is only as good as how much better it is than basic mutation type learning,  which I think actually it sounds like it wins in a logged fashion...  by a heap.  :) -- Artificial General Intelligence List: AGI Permalink:

Re: [agi] Re: A mathematics of concpetual relations?

I think it would build up slowly over time,  it cant solve problems that are too far away from what it already knows,  it has to relate to what its got in it, for more relations to develop.   So in that way it would hit "temporary local optimas" and maybe even some form of cause and effect

Re: [agi] Re: A mathematics of concpetual relations?

Steve said: I strongly suspect biological synapses are tagged in some way to only connect with other synapses carrying dimensionally compatible information. I totally agree. So one thing that I am wondering about is whether that can be computed using a novel kind of mathematics? Intuitively, I

Re: [agi] Re: A mathematics of concpetual relations?

Jim, Many systems, e.g. while adding probabilities to compute probabilities doesn't make sense; adding counts having poor significance, which can look a lot like adding probabilities, can make sense to produce a count. Where this gets confusing is in sensory fusion. Present practice is usually

Re: [agi] Re: A mathematics of concpetual relations?

I originally thought about novel computational rules. Arithmetic is not reversible because a computational result is not unique for the input operands. That makes it a type of compression. Furthermore it uses a limited set of rules. That makes it a super compression method. On Thu, Jun 20, 2019,

Re: [agi] Re: A mathematics of concpetual relations?

I guess I understand what you mean. On Thu, Jun 20, 2019, 12:07 PM Jim Bromer wrote: > I think your use of metaphors, especially metaphors that were intended to > emphasize your thoughts through exaggeration, may have confused me. Would > you explain your last post Steve? > > On Thu, Jun 20,

Re: [agi] Re: A mathematics of concpetual relations?

I think your use of metaphors, especially metaphors that were intended to emphasize your thoughts through exaggeration, may have confused me. Would you explain your last post Steve? On Thu, Jun 20, 2019, 12:02 PM Steve Richfield wrote: > Too much responding without sufficient thought. After a

Re: [agi] Re: A mathematics of concpetual relations?

Too much responding without sufficient thought. After a week of thought regarding earlier postings on this thread... Genuine computation involves manipulating numerically expressible value (e.g. 0.62), dimensionality (e.g. probability), and significance (e.g. +/- 0.1). Outputs of biological

Re: [agi] Re: A mathematics of concpetual relations?

To me a conscious mind is one that can describe (conceptualize) reality - and create reality from words. Not sure how you classify prednet as conscious. Yeah it makes and applies non-verbal concepts... Is that what you mean? Wouldn't that be called subconscious though? On Thu, Jun 20, 2019,

Re: [agi] Re: A mathematics of concpetual relations?

It has the basic structure and organization of a conscious agent, obviously it lacks the other ingredients required to produce a complete mind. Stefan Reich via AGI wrote: Prednet develops consciousness? On Wed, Jun 19, 2019, 06:51 Alan Grimes via AGI > wrote:

Re: [agi] Re: A mathematics of concpetual relations?

Prednet develops consciousness? On Wed, Jun 19, 2019, 06:51 Alan Grimes via AGI wrote: > Yay, it seems peeps are finally ready to talk about this!! =P > > Lets see if I can fool anyone into thinking I'm actually making sense by > starting with a first principles approach... > > On first

Re: [agi] Re: A mathematics of concpetual relations?

ything, wintry death is but an interlude to make space for new growth. In this event, the sooner the field is pruned, the better. Robert Benjamin From: rounce...@hotmail.com Sent: Wednesday, 19 June 2019 16:45 To: AGI Subject: Re: [agi] Re: A mathematics of concpetual r

Re: [agi] Re: A mathematics of concpetual relations?

Yay, it seems peeps are finally ready to talk about this!! =P Lets see if I can fool anyone into thinking I'm actually making sense by starting with a first principles approach... On first approximation, yer computer's memory is just a giant string of charactors. Since everything you have

Re: [agi] Re: A mathematics of concpetual relations?

the curve. Smarts alone are simply not going to cut it. https://www.cuuble.com/cuuble-mesh/ Robert Benjamin *From:* Jim Bromer *Sent:* Wednesday, 12 June 2019 15:30 *To:* AGI *Subject:* Re: [agi] Re: A mathematics of concpetual

Re: [agi] Re: A mathematics of concpetual relations?

This could be an alternative to mutation,  who wouldnt want to skip past having all that pesky trial time. -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T395236743964cb4b-Mbe272a2671c5d00ab994748b Delivery

Re: [agi] Re: A mathematics of concpetual relations?

There are feasible mathematical abstractions that we cannot imagine. There is a relevant issue - was it the Church-Turing Thesis - that supposes that any computable function can be simulated on a Turing Machine but that does not mean that we can imagine all computational methods and it certainly

Re: [agi] Re: A mathematics of concpetual relations?

If you give yourself infinite computation power an exponential brute force search of physics engine logic, contains all methods that exist in the environment, as long as it only involved still objects, which were simulated correctly,  according to some primary mover.    I was wonder if taking

Re: [agi] Re: A mathematics of concpetual relations?

I really cannot figure out what you are saying Steve. And I don't know how what you are saying could actually apply to me. "...there could be NO way to understand and/or debug such a thing," sounds like it might have some relevant meaning but then to interpret it I have to go back to, "without

Re: [agi] Re: A mathematics of concpetual relations?

it has been done? Video Compression as Fast As Possible: https://www.youtube.com/watch?v=qbGQBT2Vwvc -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T395236743964cb4b-M2417b7336fdf440be95e5249 Delivery options:

Re: [agi] Re: A mathematics of concpetual relations?

Jim, It is (nearly?) impossible to "learn" in a way that preserves value (e.g. 50%), dimensionality (e.g. probability), and probably significance (e.g. +/-10%) without constraining learning to ONLY learn things that fit this model. Without this model, it is just numerology that can NEVER EVER be

Re: [agi] Re: A mathematics of concpetual relations?

The 'formal' part of the system can be acquired through learning. Jim Bromer On Wed, Jun 12, 2019 at 9:14 PM wrote: > Yeh Steve - maybe that helps it try novel situations better??? > newsflash from me -> i think that formalizing the system manually ends up > a shallower system than what needs

Re: [agi] Re: A mathematics of concpetual relations?

Yeh Steve - maybe that helps it try novel situations better??? newsflash from me ->  i think that formalizing the system manually ends up a shallower system than what needs to be there for a developing system.    because its cheating it to do things,  is it where the term "deep learning" comes

Re: [agi] Re: A mathematics of concpetual relations?

I guess that I really don't have any good ideas for a special mathematical computational system. I want to use strings of types and specifics but they won't be mathematical and they won't be text. But the idea of keeping fundamental relations based on learned types separate from the other

Re: [agi] Re: A mathematics of concpetual relations?

Everyone seems to want to compute with positives - indicators, probabilities, confidence, etc. HOWEVER, the math is more solid when computing with negatives, e.g. the likelihood that there is NO indicator of the absence of something. The presence/absence is often difficult to compute, but the

Re: [agi] Re: A mathematics of concpetual relations?

I watched the video on Binons but not the one on Hierarchies. I have been at this for longer than I care to admit so I feel that I have seen things like this before. I am looking forward to watching the video on Hierarchies. My imagining of what I might be able to effectively use as a kind of

Re: [agi] Re: A mathematics of concpetual relations?

Programming is not boring. It can be tedious but it is like working out. When you get stronger, some work that was awful seems easy and you are glad that you can get a chance to get the exercise. When you do a lot of programming, a great deal of your mental confusion evaporates. That does not mean

Re: [agi] Re: A mathematics of concpetual relations?

Just imagine how boring it is to do the dishes.  programming it is probably more boring.  But we are grownups, we are boring, so it suits us to do it.  :) -- Artificial General Intelligence List: AGI Permalink:

Re: [agi] Re: A mathematics of concpetual relations?

You can actually set up boolean algebra that solves all sorts of tasks, making it develop the structures itself would be more closer to real agi,  but it would still be quite effective setting it up for it, old school style. -- Artificial General

Re: [agi] Re: A mathematics of concpetual relations?

The database just needs to be able to grow as it learns new things. -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T395236743964cb4b-Mfa4233300a6e7ff8aa21045a Delivery options:

Re: [agi] Re: A mathematics of concpetual relations?

Maybe the database doesnt have to be that big,  to do a lot.  I know that when one thinks of AGI he thinks of some great ginormous load of information,   but im just saying maybe small things could go really far as well. -- Artificial General Intelligence

Re: [agi] Re: A mathematics of concpetual relations?

The idea of a large mega or 100 kilo bit number, is that it allows for an extensive assignment of meaning-by-position. But only a small number of bits would be used during any calculation so a compression method would be employed. I would not actually do any calculations with million bit numbers.

Re: [agi] Re: A mathematics of concpetual relations?

This probably sounds stupid but.. With the generation idea, about getting the a.i. to develop images from an algebraic model.  It has a problem where the output is no longer symbolic (its pictures) I wonder it might be easier for the computer to describe it in words as if it was a picture...   

Re: [agi] Re: A mathematics of concpetual relations?

e 2019 02:28 To: AGI Subject: Re: [agi] Re: A mathematics of concpetual relations? I am not sure what you are saying Nanograte. I can't remember what a schema in AI is other than it might be some kind of partial definition like a set of abstractions that might be filled in with specifi

Re: [agi] Re: A mathematics of concpetual relations?

Hi Jim,  I've been following this discussion with interest. Good stuff.  If you are looking for a way of representing those millions of bits and only use a small part of it at a time and build it hierarchically may I suggest you look into binons.  In the presentations on binons and the

Re: [agi] Re: A mathematics of concpetual relations?

une 2019 19:32 > *To:* AGI > *Subject:* Re: [agi] Re: A mathematics of concpetual relations? > > Mike, > I again apologize for my strident and aggressive responses. I honestly do > not mean to be disrespectful of the remarks that have been made in this > thread.I just have a

Re: [agi] Re: A mathematics of concpetual relations?

: Tuesday, 11 June 2019 19:32 To: AGI Subject: Re: [agi] Re: A mathematics of concpetual relations? Mike, I again apologize for my strident and aggressive responses. I honestly do not mean to be disrespectful of the remarks that have been made in this thread.I just have a way of expressing myself which

Re: [agi] Re: A mathematics of concpetual relations?

Mike, I again apologize for my strident and aggressive responses. I honestly do not mean to be disrespectful of the remarks that have been made in this thread.I just have a way of expressing myself which comes across as a little rough at times. Let me give you an example of something that I just

Re: [agi] Re: A mathematics of concpetual relations?

mean't "wasn't being dismissive". Typing too fast On 6/11/19, Mike Archbold wrote: > Jim, It just reminded me of the Gardenfors work -- I was being at all > dismissive of your posts. I was just pointing the work out in case you > were not familiar with it. On the whole, I'm not dismissive of >

Re: [agi] Re: A mathematics of concpetual relations?

Jim, It just reminded me of the Gardenfors work -- I was being at all dismissive of your posts. I was just pointing the work out in case you were not familiar with it. On the whole, I'm not dismissive of anybody's ideas in AGI. It's all a wide open space IMO. On 6/11/19, Jim Bromer wrote: > I

Re: [agi] Re: A mathematics of concpetual relations?

I remember that someone kept dismissing my notion of conceptual relativism and finally he mentioned some book that had been written 40 or 50 years ago which had mentioned that concepts were relative. I wondered - could it be true? Could someone have examined conceptual relativism decades ago? I

Re: [agi] Re: A mathematics of concpetual relations?

The statement that unsupervised GAN's are the way of the real brain is an absurdity. Human beings do not yet know how the brain works. I do believe that the mind must have some way to project objects (like images of objects) onto other objects (like other images of objects or within an object

Re: [agi] Re: A mathematics of concpetual relations?

This topic reminds me of this book from almost 20 years ago: https://mitpress.mit.edu/books/conceptual-spaces On 6/11/19, keghnf...@gmail.com wrote: >  Generative Neural Networks, GAN. >  This give give a relation from stating image or data to another. > > Latent Space Human Face Synthesis |

Re: [agi] Re: A mathematics of concpetual relations?

 Generative Neural Networks, GAN.   This give give a relation from stating image or data to another. Latent Space Human Face Synthesis | Two Minute Papers #191: https://www.youtube.com/watch?v=aR6M0MQBo2w   A  programmer select two images or data points.   A  programmer put in 50 percent value

Re: [agi] Re: A mathematics of concpetual relations?

I'm not really a mathematician, but I can think outside the box about the problem. I am just thinking about an algorithm, a sub-program, that can deal with different levels of generalizations which will include collections of objects. I am not going to try to make it solve all AI problems but it

Re: [agi] Re: A mathematics of concpetual relations?

I can understand what your saying to a degree. Indeed if you trained a label "cat jumped out of the box" directly, its not the same as understanding cat, jump and box separately - which is alot more successful/invarient/reusable? The way you say this a "mathematics" problem, was alien to me,

Re: [agi] Re: A mathematics of concpetual relations?

 ANN or CNN have near or better able to detect objects, 2019.   That is 99. 98 percent of humans.  They are called black boxes because no one has a math formal to explain how they do it so well.  These balk boxes are trained on massive amounts of data.   All had crafted formula and equations,

Re: [agi] Re: A mathematics of concpetual relations?

Convincing other people to accept or use (or even think about) the mathematics that I am thinking about is not one of my goals. While my ideas are sketchy at best, they would only be useful to me if I actually could use them. Getting people to comment helps me to think outside of my own personal

Re: [agi] Re: A mathematics of concpetual relations?

Researchers try to recreate human-like thinking in machines: https://techxplore.com/news/2019-05-recreate-human-like-machines.html -- Artificial General Intelligence List: AGI Permalink:

Re: [agi] Re: A mathematics of concpetual relations?

Infusing machine learning models with inductive biases to capture human behavior: https://techxplore.com/news/2019-06-infusing-machine-inductive-biases-capture.html -- Artificial General Intelligence List: AGI Permalink:

Re: [agi] Re: A mathematics of concpetual relations?

Thats really interesting Mr. Richfield, id love to hear more about it and what it could be clues for.   Stuff working in the field (Like open ai) is still just based apon dumb mutation, they have this novelty search that would entail the robot doing everything just because its new.  If that was

Re: [agi] Re: A mathematics of concpetual relations?

 The ultimate paper of what you speakuth of are time:  https://www.academia.edu/37275998/A_Nice_Artificial_General_Intelligence_How_To_Make_A_Nice_Artificial_General_Intelligence -- Artificial General Intelligence List: AGI Permalink:

Re: [agi] Re: A mathematics of concpetual relations?

I know. If that happened squares would take me seriously. Jim Bromer On Wed, Jun 5, 2019 at 3:11 PM wrote: > ha, watch out, theories can catch on whether you like it or not. :) > *Artificial General Intelligence List * > / AGI / see discussions

Re: [agi] Re: A mathematics of concpetual relations?

ha, watch out, theories can catch on whether you like it or not. :) -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T395236743964cb4b-Md569c6d0817b7e97204f890a Delivery options:

Re: [agi] Re: A mathematics of concpetual relations?

It is unlikely that many people are going about it in just the same way as I am. Jim Bromer On Wed, Jun 5, 2019 at 2:25 PM Jim Bromer wrote: > When someone talks about deep learning they are talking about some kind of > method that 'connects' weights and other functions derived from the

Re: [agi] Re: A mathematics of concpetual relations?

When someone talks about deep learning they are talking about some kind of method that 'connects' weights and other functions derived from the weights all of which are calculated using standard arithmetic. Addition, multiplication.. I am not dismissing those ideas, in fact, I want to think about

Re: [agi] Re: A mathematics of concpetual relations?

I did not mean to be overly critical in my last email. I was just trying to be objective, to the best of my ability. A major underlying goal in discrete AI has been to look for a simple set of rules and discrete object types which could be used as a basis for all knowledge. I believe these goals

Re: [agi] Re: A mathematics of concpetual relations?

Yes keghn,  you go to maximize the function, is the current going style. -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T395236743964cb4b-M1b5e07b1ccb6f78b3dd931fb Delivery options:

Re: [agi] Re: A mathematics of concpetual relations?

 I say use  the singularity concept and measured and record use of gradient of descent to make data point same. And then go one step beyond on how to make all to act a unified as a singular being. -- Artificial General Intelligence List: AGI Permalink:

Re: [agi] Re: A mathematics of concpetual relations?

Ok, sorry about that misunderstanding. You need to form the model of the of the working area of the a.i. ,  its true for us all,  but how you do it, is what makes what your doing original. -- Artificial General Intelligence List: AGI Permalink:

Re: [agi] Re: A mathematics of concpetual relations?

I am again not sure what you are saying There are reasons to use conceptual components. They can be expressed with generalized symbols (like words) which can be narrowed down using other generalized symbols. This means that you do not have to have a separate name for everything. Using general

Re: [agi] Re: A mathematics of concpetual relations?

I dont want to stop this amazing imagination of yours.   But I must warn you ockhams razor is there to dismiss ideas on the outset,   and maybe simplistic weighting is all there is to it.   You dont always need a factor, when the symbol itself already deals out the full picture.

Re: [agi] Re: A mathematics of concpetual relations?

No I am not thinking of purely Boolean algebra. Even if it is entirely consistent with logic, (Boolean logic) it still could be different than logic. And "strength" or magnitude of some characteristic is not the only possible weighting that is possible. For example, a collection of statements

Re: [agi] Re: A mathematics of concpetual relations?

I mean you can make a symbolic system without factors,  but I guess you could have factors but it would be in the form of a weighting of a symbol, like "strength=6"  but without the 6, its purely only boolean algebra,  not algebra.  And it works without the 6 too,  is all I mean.

Re: [agi] Re: A mathematics of concpetual relations?

I am not sure what it is that you (rouncer81) are saying. Can you give a little more detail to your statement? Jim Bromer On Thu, May 30, 2019 at 8:40 AM wrote: > factors are important, but symbology has a strength where its known to be > true without detail. > *Artificial General