Manticore wrote:

Way too much ganja.

Do you have a simpler/more correct sketch of ambient isotopy?

ProgrammingGodJordan wrote:

twistor59 wrote:I downloaded the supermanifold hypothesis paper from academia.edu. It contains an abstract, summary and references section, but no body.

Can you give an example of how you use Grassmann-valued coordinates in datasets in an ML context?

That is an optimal question.

(A)

Code: Select all

The supermanifold hypothesis in deep learning prescribes the clamping of the Grassmannian parameters in a particular regime, after which largely real numbers are usable... (In other words, some grasmannian bound properties are perhaps feasible, whence observations in deep neural models don't strictly require grassmann aligned numbers)

★★ Feasible properties lay in the boundary of 'eta' ★★, or direct numerical simulations etc, at least for an initial 'trivial' example of reinforcement like learning in this paradigm.

Other references: https://ncatlab.org/nlab/show/Euclidean+supermanifold


(B)
The body is actually present:




Essentially, the neural bound supermanifold hypothesis' equation in summary, merely represents a consistency of particular layer-wise properties (homeomorphisms, bijective inverses, etc.) par input data transformation (pertinently, as it relates to some temporal difference paradigm). Norm calculations etc. offset the aforesaid properties' consistency.


(C)
(i) Think of my supermanifold hypothesis in deep learning as merely a reachable way to organize manifolds, as it relates to some temporal difference horizon.

(ii) The expression is that causal laws of physics are encodable such that these may compound in the aforesaid temporal difference paradigm.

Due to (ii), a superfield description emerges.



On the above rendition, imagine some superior identity sequence.

Looking at point B, it would be good if you can explain this a bit more: I have only a very sketchy understanding of ML:
phi is the homeomorphism - the transformation of the space of data values?
what are w and theta? I'm familiar with the perceptron where w is traditionally the weight vector

Jordan, I guess you do have some obscure esoteric mathematical knowledge pertaining to AI, but it's not accessible to most, and this forum is populated by non-mathematicians mainly. If you could explain the concepts in the OP to a community of non-mathematicians that would ingratiate you with me, at least, if not many others. I find AI fascinating but would rather you communicated on my level than have to learn the difficult math. All your esoterica does is alienate.

Very interesting and provocative OP, ProgrammingGodJordan. Judging by the comments you received so far I think it might be helpful if you go a bit slower and provide more detail. ( What you might call: adapt the learning rate. )

You suggest:

This is 'simply' because I detected evidence that life's meaning probably occurs on the horizon of optimization (Jeremy England “Dissipative Adaptation”…),

I would phrase it a bit differently: life's meaning can be observed at the horizon of optimization. The horizon of optimization defining that meaning. But why quibble over words.

Perhaps it would help the members to learn a bit more about Jeremey England's work.

https://www.youtube.com/watch?v=e91D5UAz-f4

Great thread so far...

LucidFlight wrote:Great thread so far...

I'm thinking a bit of Deep Learning would go great with that popcorn.

Keep It Real wrote:Jordan, I guess you do have some obscure esoteric mathematical knowledge pertaining to AI, but it's not accessible to most, and this forum is populated by non-mathematicians mainly. If you could explain the concepts in the OP to a community of non-mathematicians that would ingratiate you with me, at least, if not many others. I find AI fascinating but would rather you communicated on my level than have to learn the difficult math. All your esoterica does is alienate.

Hmm. God - check. Quantum mechanics - check. Reproduced snippets out of context that do not relate to one another - check.

Is that how experts communicate? Or how quantum woo functions?

ProgrammingGodJordan wrote:I am a casual body builder/...This is 'simply' because I detected evidence that life's meaning probably occurs on the horizon of optimization...

I may as well say that I have "detected" by means I won't reveal (because I'm far too smart to communicate effectively or normally) that the meaning of life resides in the tangent Minkowski space containing vectors in the space between bullshit and the zero gravity limit of the curved space Pseudo-Riemannian manifold containing tensor and so start a cult and tell everyone they're like a four year old in comparison to me.

I'm hoping ProgrammingGodJordan will help me create a deep learning generative model for blues licks - and it would be if it was a quantum model.

But even a deep learning neural net model that classifies chords as minor, major, and 7th would be interesting- and perhaps better than the classifier I have.

Keep It Real wrote:Jordan, I guess you do have some obscure esoteric mathematical knowledge pertaining to AI, but it's not accessible to most, and this forum is populated by non-mathematicians mainly. If you could explain the concepts in the OP to a community of non-mathematicians that would ingratiate you with me, at least, if not many others. I find AI fascinating but would rather you communicated on my level than have to learn the difficult math. All your esoterica does is alienate.

As a lazy atheist/coder, I am not equipped with any such "obscure esoteric mathematical knowledge".

I am attempting to sensibly confluence several sequences of paradigms, concerning my understanding of quantum computing, modern machine learning, and some empirical data, together with some theory.

Here is a modified quote that summarizes the super-manifold hypothesis in deep learning:

ProgrammingGodJordan wrote:

Essentially, the neural bound supermanifold hypothesis' equation in summary, merely represents a consistency of particular layer-wise properties (homeomorphisms, bijective inverses, etc.) par input data transformation (pertinently, as it relates to some temporal difference paradigm). Norm calculations etc. offset the aforesaid properties' consistency, pertinently, abound some parametric oscillation paradigm, containing Zλ.

The underlined section in green above, is the modification to the old quote.

The supermanifold hypothesis in deep learning, is merely manifold operation in some Zλ subsuming oscillation paradigm.



EXPLANATION:
The supermanifold hypothesis in deep learning, "simply" expresses some layer-wise transformation of some input data, existing such that we maintain certain properties between such layers, in some particular oscillation paradigm, consisting of Zλ.

The old quote discussed largely, space complexity, and little about relevant time. The Zλ portion crucially identifies why the supermanifold hypothesis in deep learning involves super manifolds.

You may find useful data here: https://arxiv.org/abs/0705.1134

For a simple picture, from which you can probably code something, See: https://www.quora.com/What-is-the-Manif ... -Bennett-9

Might we call this Transgressing the Boundaries: Towards a Transformative Hermeneutics of Quantum Gravity Computing?

ProgrammingGodJordan wrote:

Keep It Real wrote:Jordan, I guess you do have some obscure esoteric mathematical knowledge pertaining to AI, but it's not accessible to most, and this forum is populated by non-mathematicians mainly. If you could explain the concepts in the OP to a community of non-mathematicians that would ingratiate you with me, at least, if not many others. I find AI fascinating but would rather you communicated on my level than have to learn the difficult math. All your esoterica does is alienate.

As a lazy atheist/coder, I am not equipped with any such "obscure esoteric mathematical knowledge".

ProgrammingGodJordan, what programing environment does "a lazy atheist/coder" find optimal for deep learning projects?

I'm considering Keras - is that a good choice?

This looks like a good demonstration of how to use deep learning. It seems like a practical guide to how one could get started if they wanted to experiment.



It's not exactly obvious how to tie this into Jeremy England's work??? Maybe if I sleep on it.

John Platko wrote:

ProgrammingGodJordan wrote:

Keep It Real wrote:Jordan, I guess you do have some obscure esoteric mathematical knowledge pertaining to AI, but it's not accessible to most, and this forum is populated by non-mathematicians mainly. If you could explain the concepts in the OP to a community of non-mathematicians that would ingratiate you with me, at least, if not many others. I find AI fascinating but would rather you communicated on my level than have to learn the difficult math. All your esoterica does is alienate.

As a lazy atheist/coder, I am not equipped with any such "obscure esoteric mathematical knowledge".

ProgrammingGodJordan, what programing environment does "a lazy atheist/coder" find optimal for deep learning projects?

I'm considering Keras - is that a good choice?

I am mostly familiar with mxnet.

(I once used mxnet to do heart irregularty detection, using residual neural networks, in a kaggle competition. (Interesting note: My model presentation, or specifically, the result sequence created by the model, was ranked at 76/500+ in the world, near the end of the competition))

I am now experimenting with tensorflow.

Note: If you weren't aware before, these models require strong gpu(s).



(B)
I detect that like mxnet, keras abstracts away all the deep stuff.
So you can quickly put together some model, but you won't really know what is going on under neath.

As a start, I learnt how to encode a neural network from scratch: https://github.com/JordanMicahBennett/S ... -SENTIENCE

Other details:
(1) "Deep Learning Book" : http://www.deeplearningbook.org/(Includes detailed guidance by one of the fathers of deep learning Bengio)

(2) "Data science from scratch" : http://shop.oreilly.com/product/0636920033400.do (Very practical book, and easier to parse than deep learning book)


NOTE: (2) is easier to understand, but (1) contains a more wholesome picture of the grand image that is the deep learning scenario.



(C)
The following roster of deep learning libraries are useful:

https://github.com/zer0n/deepframeworks

https://www.youtube.com/watch?v=qv6UVOQ0F44&t=2s
For those wanting to enter deep neural learning, the above video may or may not generate intrigue.

ProgrammingGodJordan wrote:

John Platko wrote:

ProgrammingGodJordan wrote:

Keep It Real wrote:Jordan, I guess you do have some obscure esoteric mathematical knowledge pertaining to AI, but it's not accessible to most, and this forum is populated by non-mathematicians mainly. If you could explain the concepts in the OP to a community of non-mathematicians that would ingratiate you with me, at least, if not many others. I find AI fascinating but would rather you communicated on my level than have to learn the difficult math. All your esoterica does is alienate.

As a lazy atheist/coder, I am not equipped with any such "obscure esoteric mathematical knowledge".

ProgrammingGodJordan, what programing environment does "a lazy atheist/coder" find optimal for deep learning projects?

I'm considering Keras - is that a good choice?

I am mostly familiar with mxnet.

(I once used mxnet to do heart irregularty detection, using residual neural networks, in a kaggle competition. (Interesting note: My model presentation, or specifically, the result sequence created by the model, was ranked at 76/500+ in the world, near the end of the competition))

I am now experimenting with tensorflow.

Note: If you weren't aware before, these models require strong gpu(s).



(B)
I detect that like mxnet, keras abstracts away all the deep stuff.
So you can quickly put together some model, but you won't really know what is going on under neath.

As a start, I learnt how to encode a neural network from scratch: https://github.com/JordanMicahBennett/S ... -SENTIENCE

Other details:
(1) "Deep Learning Book" : http://www.deeplearningbook.org/(Includes detailed guidance by one of the fathers of deep learning Bengio)

(2) "Data science from scratch" : http://shop.oreilly.com/product/0636920033400.do (Very practical book, and easier to parse than deep learning book)


NOTE: (2) is easier to understand, but (1) contains a more wholesome picture of the grand image that is the deep learning scenario.



(C)
The following roster of deep learning libraries are useful:

https://github.com/zer0n/deepframeworks

Thanks for the info. The deep learning book is very hot. It has 7 holds on 2 copies in my library network, I had an easier time getting the Rolling Stones box set. I'll check out the rest of your links. It looks like Keras has a CPU only version. I'm not sure I want to program an algorithm from scratch - although I understand how there are advantages going that route.

John Platko wrote:

ProgrammingGodJordan wrote:

John Platko wrote:

Thanks for the info. The deep learning book is very hot. It has 7 holds on 2 copies in my library network, I had an easier time getting the Rolling Stones box set. I'll check out the rest of your links. It looks like Keras has a CPU only version. I'm not sure I want to program an algorithm from scratch - although I understand how there are advantages going that route.

(1)
Yes, hugely beneficial, novel works can stem from keras like level libraries.
For example, residual neural networks can be composed using high level libraries like mxnet or keras or tensorflow.

(2)
Yes, Deep Learning Book is powerful.
It is no surprise that it is in demand, as it is presented by Yoshua Bengio, a powerful figure in deep learning.
His work partly inspired the supermanifold hypothesis in the original post.

Your OP is abstruse to say the least, it seems almost deliberately so. It doesn't help that you use the phrase "meaning of life", which we all know is meaningless, and follow it up with "probably occurs on the horizon of optimization". Is this a term of art I'm too obtuse to understand or is it of the nature of a 'fitness horizon', which is a straight forward idea, understandable by almost anyone?

crank wrote:Your OP is abstruse to say the least, it seems almost deliberately so. It doesn't help that you use the phrase "meaning of life", which we all know is meaningless, and follow it up with "probably occurs on the horizon of optimization". Is this a term of art I'm too obtuse to understand or is it of the nature of a 'fitness horizon', which is a straight forward idea, understandable by almost anyone?

To some, the original post is "abstruse", to some it is trivial.

Anyway, life's goal state probably occurs on the boundary of optimization. (As far as science indicates)

Both life and "non-life" tend to approach some paradigm that optimizes some context. (Dissipative Adapatation/Jeremy ....)

So, life's goal state is likely such that matter is optimizing some context.

Life's goal state? Can you define? Also, define 'horizon of optimization'. Are these terms of art? If so, they should be easy to define, if not, it's deliberately abstruse. When someone asks for clarification, and the response includes a repeat of the term in question, and more verbiage even less transparent, it's hard not to conclude deliberate abstruseness. Surprising something so trivial as that could be misunderstood.

crank wrote:Life's goal state? Can you define? Also, define 'horizon of optimization'. Are these terms of art? If so, they should be easy to define, if not, it's deliberately abstruse. When someone asks for clarification, and the response includes a repeat of the term in question, and more verbiage even less transparent, it's hard not to conclude deliberate abstruseness. Surprising something so trivial as that could be misunderstood.

I've been trying to sus out this word choice myself.

We have several places where the OP credits Jeremy England but I couldn't find any place where
Jeremy England said that:

from

(i)
Life's meaning probably occurs on the horizon of optimization:

(source: mit physicist, Jeremy England proposes new meaning of life)

from

Life's meaning probably occurs on the horizon of optimization:

(source: mit physicist, Jeremy England proposes new meaning of life)

I do find it to be a rather insightful statement. Although like I stated, I would phrase it a bit differently. I could explain why in great detail. I'd be very interested to know where the exact words came from. It's very curious.