Replacing Science with Pattern Study

[Anonymous No.9227147]

How long before we finally get rid of science and adopt something superior?

I propose Pattern Study as the replacement. Pattern Study involves creating pattern structures like Association, Causation, up to very complex compositional patterns and then matching them with the data, in a recursive way such that the pattern structure changes to fit.

Anyway just some random idea for finally getting rid of the shitstain on humanity that is "science".

[Anonymous No.9227152]

we could also just call it general intel instead and give it some drives.

[Anonymous No.9227166]

>>9227147
Are you seriously suggesting over fitting models to the data as a replacement for science? What happens when we encounter new data from ranges not yet considered

[Anonymous No.9227511]

>>9227166
Science is based on the need of certain circumstances. It's a tool specific for humans with a certain level of technology to judge and assess the validity of things.

As technology advances and as it is now it makes sense to stop using shitty tools

[Anonymous No.9227651]

>>9227147
>getting rid of the shitstain on humanity that is "science".
Go use your own independent non-science derived international computer network to shitpost on then. And while you're at it throw away everything useful in your home that you don't think about it because you're too much of a brainlet to appreciate all the scientific innovations you would never be able to even begin reproducing yourself.
>>9227511
>As technology advances and as it is now it makes sense to stop using shitty tools
Machine learning IS the shitty tool. You only use it when you don't have the ability to understand the problem well enough to directly solve it and/or when the problem involves a massively intractable amount of trivial details that make explicit rules based programming more trouble than it's worth, like with image recognition.
You would notably NOT use ML to try to reinvent physics for example because analytical approaches have already given us extremely efficient and direct methods for modeling the motion and behavior of physical phenomena. And that's what you want whenever possible, analytical solutions based on actual understanding of the problems you're dealing with. Because the alternative is throwing all your data at nets and then getting back programs that can generate output similar to what you training it on without giving you any sort of insight into how the problem works. This is great if your goal is to build Google Translate because of course you don't want to have to program in every imaginable inane rule for translating French into Japanese. But that is a completely retarded approach to solving abstract science questions because you don't want a machine that can reproduce output, you want an understanding of the problem you're working on.

[Anonymous No.9227666]

>>9227147
You just described mathematics.

Mathematica = Isolation Filter

[Anonymous No.9228224]

>>9227651
No one understands it though. We just have simple patterns like F=MA that we then create random meanings around until they fit. None of those meanings actually work though and have to be "oh yeah that's actually all wrong" when you change the circumstances or scale.

[Anonymous No.9229641]

>>9227147
Most disgusting thing i've read in a while.

[Anonymous No.9229858]

>>9229641
Do you understand your associative links in your brain? Smells that you have linked with activities or concepts?

Now apply discernment function to these to determine causal links. It's all a simplification/compression thing and pattern matching. Just view science as a weak methodology to do such activities outside the brain. Like 56k modem connection or passenger pigeon communication, it's not necessarily ideal but was the best available way. When you can communicate more complete thoughts or complex information easily a lot of science is no longer needed. Science is like a virtual brain machine that is shared between humans.

[Anonymous No.9229926]

>>9227147
No replacement but automation, definitely, in information collection/organisation and in papers (perhaps replace papers with a different system). It can't be replaced (not yet), fucking piling up neural networks won't do it. Most useful bits of "science" come from the intuition of an individual, combined with the established rigour of academia. The technology in question is not advanced enough to progress our bodies of knowledge.

[Anonymous No.9230054]

>>9229858
>Do you understand your associative links in your brain? Smells that you have linked with activities or concepts?
>Now apply discernment function to these to determine causal links. It's all a simplification/compression thing and pattern matching.
Do you understand that "function" is a convoluted mess of billions of years of evolutionary history and not some single clearly defined operation you can just magically program after a couple years of trial and error? Machine learning is great but it isn't anywhere close to a substitute for deliberate analysis by a self-aware biological brain having person.
There isn't any law of physics that prevents a program from becoming equivalent to that, but technically not being impossible isn't the same thing as having anywhere close to the means to accomplishing that thing in question in practice. And as long as programs aren't anywhere close to the equivalent of deliberate analysis by a self-aware biological brain having person, they should be used in pretty much the way they're already being used, as solutions for topics like image recognition or self-driving cars where the desired behavior can be reproduced through error minimizing optimization problems.
As a general rule of thumb, if your idea is so difficult to accomplish that it literally requires creating artificial equivalents to full blown self-aware analytic cognition, then there's probably an awful lot of way less difficult solutions with way more favorable ROI profiles that you're missing by jumping to that idea.

[Anonymous No.9230564]

>>9230054
I invested maybe 45 minutes of thought into this total.

Let's say you use "science" all you are doing is packaging a way to look at data and pattern recognize over it.

>do experiment Y
>you will see pattern Z

That's all it is, a way to spread a pattern recognition and agree on what patterns exist. It's stupid to not simplify and compress it down to simple patternology.

On the topic of intuition. Why do you think it's not your internal set of pattern structures attempting to fit themselves to the data? Who guessed the structure of an atom or quantum mechanics before data results came back?

Yes, we can humanize it and make it seem like something special just like we can humanize a splash of paint on a canvas to be something it's not. Is that really beneficial when it leads to people being able to abuse the tools and obfuscation of science vs a simplification?

Is it not a wonder people don't understand how intelligence or AGI would work who also happen to be scientists? Does it not occur to anyone that perhaps the perspective on reality the average human has is a complete pile of bullshit?

[/psy/mon|0-X's|/sci/mon No.9230574]

>>9230564
Would Trump pass...
A) Turing Test
B) Chatbot filter
C) Spam filter
D) Contractual obligations
E) Your walls of physical intimacy
F) Your wallet
G) None of the above

[/psy/mon|0-X's|/sci/mon No.9230577]

>>9230564
Adopt two children called X and Y

Log all their activities via experiment R, grouped by equation P, against constant C.

Infer pattern Z from your chosen interval of the above.

The equation former is usually the variable {u, i, n} and you infer the value of X+Y to self via induction.

[Anonymous No.9230616]

>>9227147
Isn't science just a method to perform pattern study. How long before we stop using tools and start using superior tools? Answer: Never. To adopt the use of superior tools means we didn't stop using tools to begin with.

With any "pattern study" method, you still run the risk of having too few data points, poorly measured data points, and over-fitting as >>9227166 pointed out.

How do you know your method won't be like a paranoid schizophrenic finding patterns where there aren't any?

[Anonymous No.9230677]

>>9230574

Fake news media can't even tell Trump from a spambot. Sad!

[Anonymous No.9231080]

>>9227147
This is like putting the cart before the horse

[Anonymous No.9231286]

>>9230616
statistics

Let's assume you break it up into the two separate aspects - Searching for Pattern Structure and Function Data.

You could then define math as Pattern Structure study. Computer Science as the creation of Function Data systems. etc

[Anonymous No.9231294]

>>9231286
Also the whole entire point of this thread is to point out the inherent flaws in human thinking due to poor primitives and foundations.

Association which is the most common pattern structure isn't something we actively think about much. Meanwhile if you see a wheel you think car. There is a very serious problem when something like a general intelligence algorithm is hard to create for humanity.

[Anonymous No.9231308]

Afaik, the scientific method is a philosophy, how is this exactly outside the realm of science?

[Anonymous No.9231320]

>>9231308
minutiae

Why would you not want to adopt a better foundation? Why would humanity not want to explore these questions?

The point is that Science obfuscates more than it should which leads to errors all throughout the system and gives more leeway to exploit the tools.

[Anonymous No.9231325]

AI will be created as soon as we can integrate problem with answer.

“If we can really understand the problem, the answer will come out of it, because the answer is not separate from the problem.”
— Jiddu Krishnamurti

[Anonymous No.9231345]

>>9227147
We already do that.
Its called the scientific method :^)

[Anonymous No.9231360]

>>9231294
>There is a very serious problem when something like a general intelligence algorithm is hard to create for humanity.
The explanation is it's not the rest of humanity that's failing to do something obvious, it's you who's failing to properly understand the scope and concrete details of the concepts you're trying to discuss. You think it's an easy and obvious thing to do because your own mental model of it is so small and lacking in substance.

[Anonymous No.9231362]

>>9227147
Vulcan detected

[Anonymous No.9231416]

>>9228224
if the subtext of your argument is that inductive reasoning is bad, suck it up

[Anonymous No.9231457]

>>9231345
Yes, but an example: The best martial art can also be the one that subtracts as many things as possible.

The point is to analyze foundation rather than the peripheral. I would argue that the scientific method and science are not a good foundation. Because of their complexity and poor basic structure it leads to errors throughout the entire scientific world.

[Anonymous No.9231487]

>>9231320
>>9231457

*settles in for an entertaining thread*

[Anonymous No.9232028]

>>9231416
all humanistic reasoning is bad. all objective observational reasoning is good.

Open ended philosophical Why? questions are braindead.

[Anonymous No.9232036]

(Science minus humanisms): comes out to being far better.

[Anonymous No.9232077]

You can't fit some probabilistic model like a neural network and discover the single exception to a principle that revolutionizes understanding because of the stochastic nature pattern recognition. Outliers can't be unfalsifiably identified as true without traditional hard science

[Anonymous No.9232082]

>>9232077
Unambiguously*

[Anonymous No.9232816]

>>9231457
I agree
succinctness, purity, efficiency are desirable qualities.
The main issue with the scientific method is everyone thinks its deductive, when its clearly inductive. It can only work within itself. So what works independently from method? Until we can turn subjective into objective we cannot apply the same truth to findings, or we can, it just needs to be noted as truly subjective.
Like a false true is merely a true false true and a true false is just a true true false. Remains in a superposition until it is relatively transposed.
Consciousness is an invisible sponge, if we could distinguish the Noumena without the observer, we'd lack the need for phenomena and the drive for knowledge would cease.
We have to be just on the cusp of anything to have any progressive pull. It can be done knowingly, but ignorance is currently a powerful tool -> unam sanctum.

[Anonymous No.9233319]

>>9232816
>he hasn't transcended subjectivity-objectivity

[Anonymous No.9235102]

>>9227166
isn't that whole "we didn't consider this we must fix our model" idea the whole driving engine behind the scientific method? lol

[Anonymous No.9235416]

>>9235102
You mean is the whole "having self-awareness" idea the driving engine behind the scientific method? It's a pretty important part, yes.

[Anonymous No.9235417]

>>9227147
>creating pattern structures like Causation
Nice thought OP. So, how much correlation must be accumulated before we can equate it with causation?

[Anonymous No.9235524]

>a whole bunch of non scientists argues for a post-scientific method because they're brainlets and think x or y is harmful, or they aren't
(cigarettes, vaccines, etc)
it's time to go back to your idealism containment threads brainlets