[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/sci/ - Science & Math


View post   

File: 395 KB, 1000x1500, 1521960620670.jpg [View same] [iqdb] [saucenao] [google]
9768063 No.9768063 [Reply] [Original]

I use thought vectors visualization for everything nowadays. First imagine a 3D space with each direction being a unique variable. For instance any magnitude along the X axis is "height", A vector direction of (1,1,1) could be "cat" etc. Then any thought I have is represented by a 3d polygon with various shapes depending on what it is. Through this analysis all creativity can be achieved by simple reshaping, add the points together to achieve something similar to outer product, etc.

Do you use thought vectors?
Are you familiar with how we can use them to enhance creativity and understand systems via vector manipulation?

I use a simple Geometric algebra style understanding to analyze thoughts in 3D space, converting to arbitrary dimension for Algebraic manipulation then back to 3D for visualization.

>> No.9768067
File: 26 KB, 245x314, what?.jpg [View same] [iqdb] [saucenao] [google]
9768067

what?

>> No.9768075

For instance some thought can be visualized by a 3d polygon of some insane shape. Then convert it to arbitrary binary vector, manipulate with simple outer product, contraction, etc, then back to 3D for visualization.

If analyzing something like "science" in this way I can learn about how it works and other similar systems. For instance the constructive nature of some things versus observational nature. Whereby science doesn't really properly categorize things based on observational deduction versus constructive analysis based on some more elementary framework. It seems to me like the systems in science all are by dumb people who don't analyze these things or can't comprehend the nature of these distinctions properly.

For instance setting these frameworks seems to be done automatically rather then a deliberate system of analysis. If you analyzed physics for instance you would see all sorts of "floors" of which elementary constructs branch out of constructively to explain observational effects. For instance classical mechanics is on the third floor, branches out to create elementary tools to predict observational effects in said floor, whereas quantum occupies a different lower flower and is constructed to match with observational effects above it.

Yet in science these concepts and categorizations are sort of left as murky pools instead of explicitly detailed an outlined. The scientific framework and system seems to not be exposed and rather everyone occupies space inside of it instead of seeing it from a vantage point.

>> No.9768077

>>9768067

Imagine a set of systems of which science is a member. Now construct the rest of the set by altering science along certain variables it contains. For instance, imagine a "system" in the set that contains science that ONLY utilizes constructive via elements and zero observation

There are so many systems and I use so many but no one else does at least while realizing it.

>> No.9768078

Is this the platonic form of autism?

>> No.9768083

>>9768078
Thought vectors are simple

Cat [0,0,1,0]
Dog[0,0,0,1]
Fat [ 1.0,0,0]

Fat Dog [1,0,0,1]
Fat Cat [1.0,1,0]
Cat Dog [ 0, 0, 1, 1]

It's just a vector representation of any idea along a basis space.

>> No.9768086

>>9768083
So to continue

Break down science into a vector then define a set of systems similar to it that exist in some space. Constructively creat

>> No.9768125

>>9768086
Fuck me, I feel like an actual bully now.
Why not just visualize it using sets? Way more efficient.

>> No.9768133

>>9768063
looks similar to word2vec from neural networks theory
https://deeplearning4j.org/word2vec.html

>> No.9768207

>>9768133
>https://deeplearning4j.org/word2vec.html
Obviously the inspiration to some degree. Although I've always thought this way due geometric algebra background, orthogonality, outer product, contraction, etc applied to other concepts.

One such application is political analysis. For instance seeing that the vector the media is using is orthogonal to election day choices (ineffective attacks)

>> No.9768210

Holy shit I thought I was the only one doing something like this

>> No.9768212

>>9768207
The application though is in creativity space. Creativity can be seen as vector manipulation of existing ideas. So if we are analyzing science and looking for novel ideas you would manipulate said vector and see what happens.

One of the interesting things is that science is very related to intelligence. Just a sort of distribution network for intelligences to share things. Almost all the actual work is done by brains and not the scientific systems in place yet science is viewed as some system not involving general intelligence.

>> No.9768216

>>9768210
It's how our brains work.

>> No.9768234

>>9768125
>visualize it using sets
how so? The whole point is to begin to look for basis vectors for science and experiment with changing the magnitude of them or setting to on/off.

The attempt is to understand science from a higher level of abstraction.

>> No.9768253

I'll do a short train of thought example. Hopefully some people can follow.

The current system of science is basically haphazard. Huge portions are not-explicit and not-defined and rely on brains. This is also similar to how say a government is designed which I could go into the flaws of for an hour.

Let's take science for example. We have two concepts: Observation and Construction. Names arbitrary

Observation: Recorded data from observation.
Construction: Creating a predictive function from some elements that correlates/matches with observation.

You can imagine this as two basis vectors within science's space. In some cases we have entirely observational data, in other's purely constructive predictive functions (for instance untested theory)

This is sort of how an analysis of science is done via thought vector breakdown to basis vectors to understand it better.

The scientific system does contain these basis vectors but is not necessarily clear or explicit in definition to their existence. For instance a scientist might intuitively understand these concepts but not explicitly.

>> No.9768262

>>9768253
To break it down further for the low IQ people who need more SPOONFEEDING

We define these two basis vectors earlier: Observation, Construction

Science contracted with Observation -> Sci/Obs

Sci/Obs, No Observation in the system. So the system contains only prediction functions with zero observational data. The predictive functions that exist in this space are infinite, all valid functions since they always are true.

aka Mathematics.

Now. Sci contracted with Predictive functions -> Sci/Pre

Sci/pre contains no functions. Only observational data. This is basically how an inanimate camera views the world. Seemingly pointless, no reasoning / pattern matching ever applied.

>> No.9768265

Because everything is ultimately 'shape/measurement/symbolic' comparison, then yes any sufficiently simplified data point can be represented/expressed/stored as a set of 'triangle versus square' style arguments occurring within bounded rings.

Like, I wouldn't care if a superior intellect infected me with perfect reasoning skills to win any argument possible. It is gifting an advanced toolset that is easy to adopt and benefits the host immediately upon understanding, so why not continue with it?

>> No.9768271
File: 24 KB, 250x242, 250px-Coord_system_CA_0.svg.png [View same] [iqdb] [saucenao] [google]
9768271

A vector however is no different than saying you agree on the congruence of three semi-related axis.

To be smart is to know that there are 'simpler representations that hold authority greater over more complex representations'. Being able to compress your idea and expand it non-destructively represents a strong enough structure to hold virtually any idea on. Like the perfect evolutionary position to be as a person of thought is, "Fuck other people's opinion, intellectual vampirism is the way to go!" (Because whatever fucking story that sells at this point, right? Just got to find the right one that gets everyone to wake up a little.)

>> No.9768278

Geometric Reasoning -> exploration of a thought vector via deconstruction into arbitrary basis vectors and analyzing changes to them via simulation.

>> No.9768281

>>9768278
Example

Imagine all variations of what you would consider to be a catdog. You are exploring the vector space contained in cat^dog via simulation (imagination)

>> No.9768282

>>9768063
I can see a couple of flaws with this approach, the big one being that you're assuming the vector space axioms apply to whatever you're modelling with vectors

Second, you seem to have limited yourself to vectors of very low dimensionality when you could just work in abstract vectors, without even depending on a basis.

>> No.9768287
File: 24 KB, 402x402, 402px-Annulus_area.svg.png [View same] [iqdb] [saucenao] [google]
9768287

>>9768278
An argument for annulus to be better taught must be made.

Accepting a default amount of measurable non-traversable space as being the observer/perspective measured. The measurement of initial offset. Powerful stuff.

>> No.9768289

>>9768282
Well thought vectors are high dimensionality. It's simply a visualization I use for them, imagining them as 3D polygons.

>> No.9768292

>>9768287
low IQ tard

>> No.9768296

>>9768282
Also I don't use vector space axioms. It's a visualization technique. The rules governing thought vectors are case dependent and more similar to and/or/n

>> No.9768302

>>9768063
Can you state your education, I am just really interested in what kind of people come up with these ideas

>> No.9768306

>>9768292
Good example. How is this useful data? Use it as a sample set for filtration of others. It's like people just gift their 'avoid me' signatures, which are things that can be reformed.

>> No.9768313

>>9768063
I have actually considered vectors as the one formalism that seems to fully encompass the absolute madness of postmodern gender politics

It at least seems to satisfy all vector space axioms:
>Associativity and addition of genders seems to hold, they don't care about the order of their genders
>0 element, called "agender"
>Additive inverses have been claimed to exist
>Scalar multiplication seems evident from the "demigender" designation, which appears equivalent to multiplying a given gender by 0.5 or so
>Distributivity properties also seem to hold

We can even define the inner product of two genders as a measure of how gay a given pairing is.

>> No.9768318

>>9768306
Your criticisms are stupid. The thread is not buzzword jargon as you imagine. Just leave the thread fucktard.

see https://deeplearning4j.org/word2vec.html

but I'd imagine you are too braindead and stupid to learn anything new.

>> No.9768325

see: https://www.cell.com/action/showImagesData?pii=S0092-8674%2817%2930538-X

as well

>> No.9768329

>>9768318
Again the capacity to continue to parse the data is of interest. I gain more by querying what I consider to be an alien opinion, regardless of its attitude. Data is data to learn from.

>> No.9768400

>>9768063
Karma??

>> No.9768421

Wow this is sad. If you've ever wondered why no one takes ML seriously on here, this is why.