ProgrammingGodJordan

This thread concerns a discussion about why physicist need to study consciousness, and probably involve themselves in the development of artificial general intelligence.

Firstly, below is a scintillating talk on artificial general intelligence given by, Theoretical Physics PHD, and former Dwave CTO, Geordie Rose.

He left Dwave, and is now instead CEO and co-founder of Kindred Ai. 

He gives an entertaining talk in the tech-vancouver video below, about why the development of artificial general intelligence is crucial for mankind.

Extra: See also a talk here, by Suzanne Gildert here (She also left Dwave, to start Kindred Ai. She has an Quantum Physics PHD as well, and used to work on superconducting chips at Dwave)

Secondly, as Max Tegmark expressed in a youtube video here, physicists have long neglected to define the observer in much of the equations. (The observer being the intelligent agent)

Alert: Notably, when I refer to consciousness below, I echo Bengio's words from one of his recent papers, "I do not refer here to more elusive meanings that have been attributed to the word “consciousness” (like qualia (Kriegel, 2014)), sticking instead to the notion of attentive awareness in the moment, our ability to focus on information in our minds which is accessible for verbal report, reasoning, and the control of behaviour."

As far as science goes, consciousness is likely definable in terms of very complex equations, from disciplines, like Physics; as an example, degrees of general structures such as manifolds central to physics and mathematics, are now quite prevalent in the study of Deep Learning.

Footnote:

As I indicate in the content above, there are two crucial end points:

(1) Physics intends to describe the cosmos, and as Max Tegmark mentions, a non trivial portion of physics, namely the observer has long eluded physics, and so the observer's framework/consciousness (as described in the alert above) warrants non-trivial analysis/development.

(2) Understanding consciousness (as described in the alert above), may lend large help to the development of artificial general intelligence, which is often underlined as mankind's last invention, that apart from solving many human problems, (i.e. Self-taught artificial intelligence beats doctors at predicting heart attacks) may also aid in the development of physics. (i.e. Ai learns and recreates Nobel-winning physics experiment)

As Max Tegmark expressed in a youtube video here, physicists have long neglected to define the observer in much of the equations. (The observer being the intelligent agent)

Perhaps consciousness may be defined in terms of very complex equations, from disciplines, like Physics; as an example, degrees of general structures such as manifolds central to physics and mathematics, are now quite prevalent in the study of Deep Learning.

On 9/6/2017 at 3:36 PM, kirving said:

That is the regular number operational to super number operational relation. Similar to how at the end of the finite series where gamma interlacing represents the next higher order or the next infinite set, infinite becomes super number in relative operation in the link you just posted.

Your talk reminds me of Deepak Chopra.

Chopra quote: "overthrowing the climactic overthrow of the superstition of materialism".

Advice: Try not to sound like Chopra.

1 hour ago, uncool said:

That makes it sound like all you're doing is consulting your memory instead of a lookup table. It's like saying you have a shortcut by replacing 8*7 with 56, instead of looking it up. 

 

I mean, you can make it look like it's a shortcut, but...

 

That's not even getting into pedagogical issues. 

Yes, you can use memory to look on the merely three standard trig rule collapser forms. (Just like the memory you could use to memorize the many many more standard trig identities)

So, using my collapser is still cheaper than looking up the many more trig identities.

You gain shorter evaluations, and you also compute with far less lookup.

FOOTNOTE:

"Uncool", thanks for your questions. I have improved the "Clear explanation" section in the paper, and removed some distracting typos too. (In addition, in the original post, the term: "∫ (√1-sin²θ/(√cos²θ) ) · cosθ dθ". should have been: "∫ (√1-sin²θ) · cosθdθ" instead, based on the problem in the video)

1 hour ago, uncool said:

So your collapser is just...doing the same multiplication, but skipping the explicit step of multiplication?

No.

Notice this preliminary Step (1): ∫ sin²ϴ · √1-sin²ϴ · cosϴ dϴ

With my collapser, you can easily identify cosϴ from the initial substitution line; so instead of writing down the "√1-sin²ϴ" term, then finding cos²ϴ, then square rooting it, you go straight ahead and evaluate cosϴ in the integral.

As a result, you don't need to look for cosϴ from 1-sin²ϴ in the identity table, and you don't need to square root.

In the scenario above, three preliminary lines are replaced (excluding explicit multiplication) and in other problems, more preliminary lines may be replaced (also excluding explicit multiplication).

Either way, you avoid searching the identity table to begin evaluation, and you avoid square rooting.

On 9/7/2017 at 1:53 PM, uncool said:

Sorry; walk me through your "collapser" for the integral of x^2 sqrt(1 - x^2) dx, then?

Without TrigRuleCollapser:

Let x = sinϴ ⟹ dx = cosϴ dϴ

Preliminary Step (1): ∫ sin²ϴ · √1-sin²ϴ  · cosϴ dϴ

Preliminary Step (2): ∫ sin²ϴ · cosϴ  · cosϴ dϴ

Evaluation: ∫ sin²ϴ · cos²ϴ dϴ

... ...

******************

With TrigRuleCollapser:

Let x = sinϴ ⟹ dx = cosϴ dϴ

Evaluation: ∫ sin²ϴ · cos²ϴ dϴ

... ...

(No preliminary steps required, you Evaluate rule: xⁿ  · dx/dϴ · dx)

12 hours ago, Mordred said:

Good paper, a proper understanding of every line and formula could take a month. Unless your already familiar with every formula and terminology.

This is the trick, everytime you see a term or formula you don't recognize.

Stop and research that line, formula or terminology.

For example you read as per the Euler Langrene. or Hamilton. Stop and study those particular topics.

You would be amazed how much detail is in every single sentence.

Good advice.

I know it is excellent advice, because I had recently invented a framework for thought, that enforces heavy scientific scrutiny.

I know how to isolate symbols and analyse them too, because I had invented some small degree of calculus in the past.

34 minutes ago, Mordred said:

Supersymmetric groups are incredibly complex for any computer regardless of how it is programmed.

So yes you do have your work cut out for you

No wonder Ai researchers are still in the regime of euclidean space instead of euclidean superspace.

For example, here is yet another paper, concerning manifold learning - mean field theory, in the Riemannian geometry: https://arxiv.org/pdf/1606.05340v2.pdf

The intriguing paper above is manifold resource I could learn from, in order to continue the supermanifold hypothesis in deep learning.

4 hours ago, Mordred said:

Could not get the bmatrix working above but have theta 1 above theta 2 for some reason I could not use theta under bmatrix.

 

Anyways you need the above function for deep thinking. That will be up to you not I lol.

As far as symmetry groups are concerned the references you mentioned simply will not cut it unless you also understand lie algebra.

Not for what your looking for ie symmetric vs supersymmetric under physics.

(these are different particle groups) example SO(10)MSSM vs SO(10)MSM you need the subgroups etc which the Clifford algebra is the preliminary.

Well here you can see for yourself

https://www.google.ca/url?sa=t&source=web&rct=j&url=http://www.slac.stanford.edu/cgi-wrap/getdoc/slac-r-865.pdf&ved=0ahUKEwi4_cPTr5LWAhUHx2MKHc8CBHMQFggfMAE&usg=AFQjCNF0FSS9Q153eGVkavToRJuwaUfxTA

Directly no, but the mathematics behind that paper is QFT treatments. Which I do understand you may want to study Hamiltons which is "action"

Thanks for the helpful message and references.

END-NOTE:

Source a provided a spark for researching super-symmetry in a computational manner.

Source b provides one of the richest resources for deep learning, while underlining manifolds, that bare non trivial relation to source a.

While sources like a and b persist as rich sources of data usable for the task at hand, I detected that they alone would probably not suffice, given that considering supermanifolds beyond manifolds in deep learning, is as far as I could observe, novel waters.

So, I know that it is likely quite necessary to study and experiment beyond the sources I presented.

48 minutes ago, Mordred said:

At this stage were not worried about the actual groups but you do need to recognize the mathematical structures and symbology.

lets demonstrate. First Deep thinking is a subset of machine learning.

We use the function h(x,θ) where x is a vector whose vector components is a greyscale  that intensifies at each pixel.

Lets predict the number of apples grown with the number of days of rain. Let x_1 be the number of apples, x_2 the number of days of rain.

So at each datapoint  x=[x1,x2]T our goal is to have a learning model

h(x,θ) where parameter vector θ=[θ0,θ1,θ2]2

such that 

Missing argument for \binom Missing argument for \begin

(12)

Based at least, on the contents of bengio's deep learning book, I am at least knowledgeable about a good portion of the symbols, (some used in relation to superspace as seen in the OP's paper)

9 hours ago, uncool said:

Walk me through your "collapser" for the integral of sin^3(x) dx?

The formulation works for many classes of integrals whose problems constitute some square rooted distribution.

Unfortunately, I don't know whether universal ways of collapsing are possible.

49 minutes ago, Mordred said:

Ok well to start with your going to need to understand what those symmetry groups represent before you can even think about using them in any program let alone deep thinking algorithms/(programs etc).

I would start with vector symmetry of linear then angular momentum systems. This is extremely important to understand any symmetry let alone a supersymmetric group.

How much differential geometry have you got?

Assuming decent math skills I recommend Clifford algebra. It will stage you on Unitary and Orthogonal groups.

Here (this is preliminary for Lie algebra)

https://www.google.ca/url?sa=t&source=web&rct=j&url=http://www.reed.edu/physics/faculty/wheeler/documents/Miscellaneous%20Math/Clifford%20Algebra.pdf&ved=0ahUKEwif8MDgkJLWAhURx2MKHfJoAB0QFggdMAA&usg=AFQjCNE-YhE0fl70gPHZhkmGsc5H1NhXiA

 

While I can help on group theory I can't directly on "Deep thinking" a brief reading up on it indicates it can use the group functions in its programming so the link above is still useful.

As far as programming is concerned my experience is in Ladder/Relay logic as per plant automation systems and Robotics. Which isn't going to help much here as the programs I've seen appear to be object oriented C++ programs.

 

The good news is you don't need the physics to understand lie algebra (at least not for the group structures,rules,axioms etc) Thats part of the beauty behind symmetry groups...Lol truth be told understanding the groups allows a huge step to understanding any physics subject.

 

Thanks for the supportive, considerate message.

Yes, I at least know of the class of symmetry groups that are required. (Relating to the bosonic riccati)

However, do you know anything about Montroll kinks, and the degrees of freedom it affords in variations of signal energy transfer in biological brains?

FOOTNOTE:

When I said learning the laws of physics in the third response above in this thread, in particular, I was referring to the supersymmetric structure, rather than my self, much like how deepmind's manifold based early concept learner infers laws of physics, based on the input space of pixels.

Models that learn things better than humans, is typical in deep learning.

42 minutes ago, Mordred said:

You definetely need some serious work to actually demonstrate what your driving at.

Anyways lets start with your actual goal.

What are you specifically wanting to program as a Deep learning format?

Your paper doesn't particularly clarify how you plan on deep learning Euclidean symmetry groups.

What is your focus? as quite frankly you are posting numerous related topics without any focus on any particular goal

 

The short answer:

As I answered above, (and as the papers outline), the goal is to use a supermanifold structure, in a bellmanian like regime, much like how Googke Deepmind uses manifolds in their recent paper.

The longer answer:

At least, from ϕ(x;θ)Tw, or the machine learning paradigm:

In the machine learning regime, something like the following applies:

• Jordan Bennett's answer to What is the Manifold Hypothesis in Deep Learning?

FOOTNOTE:

I don't know much about supermathematics at all, but based at least, on the generalizability of manifolds and supermanifolds, together with evidence that supersymmetry applies in cognitive science, I could formulate algebra with respect to the deep learning variant of manifolds.

This means that given the nature of supermanifolds and manifolds, there is no law preventing ϕ(x;θ,)Tw, some structure in euclidean superspace that may subsume pˆdata (real valued training samples), over some temporal difference hyperplane.

10 minutes ago, Mordred said:

I'm sorry but I really cannot see how you plan on connecting the dots/references etc.

Yes babies learn, yes everyone uses physics in some form or another.

 

What does that have to do with lie algebra?

Machine learning models use some structure as their memory, in order to do representations based on some input space.

Supermathematics may be used to represent some input space, given evidence that super-symmetry applies in cognitive science.

Learning the laws of physics may be a crucial part of the aforementioned input space, or task.

Pay attention to the segments below. [12] refers to: https://arxiv.org/abs/0705.1134

This is a clear explanation w.r.t. the "Trigonometric Rule Collapser Set", that may perhaps be helpful. (See source)

The above is not to be confused for u-substitution. (See why)

In the sequence: x = sin t, where dx = cost dt, and 1 − x² = 1 − sin2 t = cos2 t ....(from problem: ∫ √1- x²)

..the novel formulation dx | dt · dx occurs such that the default way of working trigonometric equations is compressed, permitting the reduction of the cardinality of steps normally employable.

For example, in the video above, while evaluating ∫ √1- x² dx, in a preliminary step, the instructor writes ∫ (√1-sin²θ/(√cos²θ) ) · cosθdθ. 

Using my trig collapser routine, this step (which may be a set of steps for other problems) is unnecessary, because applying my trig collapser set's novel form: dx | dθ · dx, we can just go right ahead and evaluate  ∫cosθ·cosθdθ.

The trigonometric rule collapser set may be a topic that may be an avenue which may spark further studies.

On 3/11/2017 at 8:52 PM, Bignose said:

So, you 'prove' that this isn't u-substitution by using something different that what your 'collapse' does.

 

You use x=4sinθ ... why not try u=4sinθ ? And maybe not make a mistake in forming the du term.... "u = 4cosθ, and du = 4cosθdθ" is obviously wrong.

 

 

Lastly, is there any reason you can't use this forum's LaTeX capabilities? Your post here is very difficult to read and it doesn't have to be...

Yes, that was a bit confusing.

This should be a better explanation: https://drive.google.com/file/d/0B8H3Ghe4haTWbW1uVGxiZ3ZqbEk/view

This thread concerns attempts to construct artificial general intelligence, which I often underline may likely be mankind's last invention.

I am asking anybody that knows supermathematics and machine learning to pitch in the discussion below.



PART A
Back in 2016, I read somewhere that babies know some physics intuitively. 
Also, it is empirically observable that babies use that intuition to develop abstractions of knowledge, in a reinforcement learning like manner.



PART B
Now, I knew beforehand of two types of major deep learning models, that:

(1) used reinforcement learning. (Deepmind Atari q)
(2) learn laws of physics. (Uetorch)

However:

(a) Object detectors like (2) use something called pooling to gain translation invariance over objects, so that the model learns regardless of where the object in the image is positioned
(b) Instead, (1) excludes pooling, because (1) requires translation variance, in order for Q learning to apply on the changing positions of the objects in pixels.


PART C
As I result I sought a model that could deliver both translation invariance and variance at the same time, and reasonably, part of the solution was models that disentangled factors of variation, i.e. manifold learning frameworks.

I didn't stop my scientific thinking at manifold learning though.

Given that cognitive science may be used to constrain machine learning models (similar to how firms like Deepmind often use cognitive science as a boundary on the deep learning models they produce) I sought to create a disentanglable model that was as constrained by cognitive science, as far as algebra would permit.



PART D

As a result I created something called the supermanifold hypothesis in deep learning (a component in another description called 'thought curvature'). 

This was due to evidence of supersymmetry in cognitive science; I compacted machine learning related algebra for disentangling, in the regime of supermanifolds. This could be seen as an extension of manifold learning in artificial intelligence.

Given that the supermanifold hypothesis compounds ϕ(x,,)^Tw, here is an annotation of the hypothesis:

1. Deep Learning entails ϕ(x;)^Tw, that denotes the input space x, and learnt representations .
2. Deep Learning underlines that coordinates or latent spaces in the manifold framework, are learnt features/representations, or directions that are sparse configurations of coordinates.
3. Supermathematics entails (x,,), that denotes some x valued coordinate distribution, and by extension, directions that compact coordinates via , .
4. As such, the aforesaid (x,,), is subject to coordinate transformation.
5. Thereafter 1, 2, 3, 4 and supersymmetry in cognitive science, within the generalizable nature of euclidean space, reasonably effectuates ϕ(x,,)^Tw.

QUESTIONS:

Does anybody here have good knowledge of supermathematics or related field, to give any input on the above?

If so is it feasible to pursue the model I present in supermanifold hypothesis paper?

And if so, apart from the ones discussed in the paper, what type of pˆdata (training samples) do you garner warrants reasonable experiments in the regime of the model I presented?

11 hours ago, DrKrettin said:

The question of whether somebody is capable of communicating in a particular language or not can only be answered by a consensus of those trying to understand the communication. This capability cannot be classified as a fact, and in your case I'm afraid the consensus is heavily against you. 

Given the evidence, I seek to enhance how I communicate.

12 minutes ago, Strange said:

I am just focusing on the fact that his belief he can communicate in English is a false belief.

Facts may persist regardless of belief...

1 hour ago, John Cuthber said:

It depends whether you understand and act on it.

Pele is (was?) a famous very good footballer. He must have been good if I had heard of him because I have no interest in the game.The details don't matter.

Commentators are noted for hyperbole so it's likely that one of them somewhere has described Pele as a "God among footballers".

It doesn't mean that Pele created the universe.

 

Sam Harris has said that we humans would be "gods" to the robots we build- we would have power of life an death etc.

But it doesn't mean that we are the creators of the universe any more than Pele.

 

Yet that's how you seem to have taken his mention of the word "God" you have interpreted it as evidence of his belief in a real supernatural creator.

That's not what he said.

 

 

Did you miss the OP?

No where there, did I express that "we are the creators of the universe".

FOOTNOTE:

Notably, I am atheistic, and the original post likewise stipulates of Sam's atheistic nature.

1 minute ago, Ten oz said:

One big problem those of faith have when attempting to interpret philosophical concepts related to religion of athiests, like Sam Harris, have is that those who believe already have a tangible and form view of who and or what God is. It is sort of like when tin foil hat wearing alien hunters applaud scientists saying they don't think earth is the only home of life in the universe. Acknowledging life may exist elsewhere in the universe in no way shape or form supports or defends Area 51 or Phoenix lights conspiracy theories but people attempt to make those connection. 

 

In Science info is only ever at the level of the best as it is known. I believe in gravity to the best I can given the education and knowledge I have to understand it. As knowledge goes and changes so to what may understanding. Nothing is taken on faith. Those who believe in God do so without the best knowledge and education they can understand. Saying life may exist on Mars isn't equal to saying human like aliens built pyramids on mars. One is an acknowledgement of the limits of what is known and the other is wild leap. 

I apologize, for I am unable to parse your comment above.

6 minutes ago, John Cuthber said:

I look forward to ProgrammingBodJordan reading some football commentator's  description of Pele as " a God among footballers" and citing it as  "evidence" for the existence of God; the Creator.

Of what consequence is your comment above?

FOOTNOTE:

I don't watch football, and I know not who "Pele" is.

3 hours ago, Manticore said:

All Sam Harris actually said was that one day we might be able to build a computer which to us today would appear to have godlike powers. This idea dates back at least to "The Last Question" by Isaac Asimov which was first published in 1956.

Yes, I had enjoyed a majority of Asimov's publications some years ago, including the Last Question, and The Last Answer.

I recall of an intriguing short story, also concerning artificial general intelligence, called "A senseless conversation", by Zach Barnett.

FOOTNOTE:

As such, the source presented, merely underlined a simple, scientifically grounded equivalence, for the long-established impression amidst the topic of creation, predominant to religious persuasion, namely God.

Such a God, as redefined, was as I had long expressed, similar to the God Sam Harris was referring to, pertinently in the regime of artificial general intelligence. 

20 minutes ago, Strange said:

Not in English. This seems to be a (false) belief on your part. 

Why do you bother to ignore evidence?

https://www.google.com.jm/search?q=concern&oq=concern&aqs=chrome..69i57.1064j0j1&sourceid=chrome&ie=UTF-8

9 hours ago, beecee said:

The gist of the OP seems to be that any belief is essentially wrong or invalid: I find that rather weird  to put it mildly. I believe and accept the scientific method: I base that belief on observation and history. How can one lack belief in all things?        

 

No.

I clearly expressed that belief is a model, that permits both science, and non-science.

However, belief typically facilitates that people especially ignore evidence. (As research, and definitions show)

A model that permits the large ignorance of evidence contrasts science.

Instead, we may employ scientific thinking, that largely prioritizes evidence, rather than a model (i.e belief) that facilitates largely,  ignorance of evidence.

9 hours ago, Strange said:

It makes zero sense to me. Which meaning of concern (as a transitive verb) do you think you are using? 
https://www.merriam-webster.com/dictionary/concern 

None of them seem appropriate in this construct.

Or perhaps you meant "We may be highly concerned with the evidence".

To concern may be to consider.

So, an alternative is: "We may highly consider evidence.".

23 minutes ago, Itoero said:

If there is a god then it's not a personal god but then it's imo an impersonal force. This idea that if there is a god, it should help people, is because people ascribe human traits to god...which does not make sense.

Reminds me of some words said to me in the distant past: "Maybe our misery and or joy, and our cosmos as a whole, is merely entainment, as a game much like gta, created by intelligent entities".