Name a dumber psyop.

This is my Nth "name a dumber psyop" thread this week. Still no dumber psyop named.

>>512121217
Sub-90-IQ meme attempt.

>>512121096 (OP)
You’re 100% correct, but who cares?

>>512121096 (OP)
Abrahamic Christianity will always be the dumbest psyop

>>512121096 (OP)
Our entire life is also a bunch of if-elses, choices define us. So I guess in a sense AI follows us closely.

>>512121096 (OP)
i'm still trying to come up with one

>>512121294
> be 2020
> AI can generate coherent text
> "pfft, it's just autocomplete"

> be 2024
> AI makes photorealistic images from text, generates music, writes code, and passes medical exams
> "meh, it's not *real* intelligence"

> be 2028
> AI designs a new material that solves global energy storage
> "still not AGI, lol"

> be 2032
> AI creates a self-replicating lunar colony and negotiates world peace
> "just fancy algorithms, bro"

> be 2040
> AI uploads consciousness of terminally ill humans into virtual immortality
> "wake me up when it can feel emotions"

> be 2050
> AI builds a Dyson sphere and predicts the end of the universe
> "it's just advanced pattern recognition"

>>512121531
>thirdie says human life has no value
lol every time.

>>512121096 (OP)
its actually more life

((X + Y) / 2)

and more like

+-------------------
: small_apple(2)
|
: hide(5)
|
: like(10)
: dislike(30)
+--------------------------
: big_apple(3)
|
: hide(10)
|
: like(10)
: dislike(30)
+------------------------
: small_apple(2)
: big_apple(3)
|
: hide(5)
: hide(10)
|
: like(10)
: dislike(30)
+-----------------

>>512121612
Learn to read maybe? Not sure how you even came up with that conclusion...

>>512121619
where `hide` gets updated also known as `learning`

>>512121569
>> be 2020
>> AI can generate coherent text
>> "pfft, it's just autocomplete"
That's still what it is in 2025.

>>512121096 (OP)
AI doesn't work like that.

here is an ordinal table with unique names

+--+----------------+---+
| 1 | small_apple | 2 |
+--+-----------------+--+
| 2 | big_apple | 3 |
+--+-----------------+---+
| 3 | like | 10 |
+--+-----------------+----+
| 4 | dislike | 30 |
+--+------------------+----+

>>512121569
If LLM-philes being subnormal victims of science fiction needed further illustration we have this post.

>>512121531
>Our entire life is also a bunch of if-elses, choices define us. So I guess in a sense AI follows us closely.
So you go through life mindlessly and deterministically following a fixed rule-set that doesn't even exhibit Turing completeness?

>>512121096 (OP)
This retarded Moldovan spams this shitty image every day multiple times a day. I don't buy AI hype but it's just not a good representation of how current AI algorithms work.

>>512121707
>bro cars still have wheels after 100 years
Retard

>>512121907
>it's just not a good representation of how current AI algorithms work.
Why not?

>>512121907
He proves that midwits are most afraid of AI.

>>512121932
Guessing the next token is not a valid thought process. Seethe about it, golem.

>>512121907

>>512122022
Then why are you seething about it so hard if it's just autocomplete?

>>512121886
I mean I'm just saying we all make choices, so does AI, but now it's at a point where these choices aren't "fixed" and it actually goes through a fuckton of shit for the output. It might be complete bullshit or it might actually be right, just like us. It doesn't make AI more advanced than people, but it's not autocomplete-tier crap anymore...

>>512122271
Guessing the next token is not a valid thought process. Seethe about it, golem.

>>512121886
You're too retarded to realize how everything you do every day is a result of you learning from millions of if>then scenarios.
>if I touch a stove then I get burned
NOOOOOOO IM NOT LIKE AI I HAVE A SOUL NOOO

>>512122312
Can you not think either, midwit?

>>512121861
this is only part of the story though

there are exactly 3 parts of the story

1. Average / Regression to the Mean ((X / Y) / 2) that is mostly used to extract data from the big data centers and find what groups of people like and dislike

2.
+-------------------
: small_apple(2)
|
: hide(5)
|
: like(10)
: dislike(30)
+--------------------------
: big_apple(3)
|
: hide(10)
|
: like(10)
: dislike(30)
+------------------------
: small_apple(2)
: big_apple(3)
|
: hide(5)
: hide(10)
|
: like(10)
: dislike(30)
+-----------------

+--+----------------+---+
| 1 | small_apple | 2 |
+--+-----------------+--+
| 2 | big_apple | 3 |
+--+-----------------+---+
| 3 | like | 10 |
+--+-----------------+----+
| 4 | dislike | 30 |
+--+------------------+----+

this part is used for the `training part`

3. but there is a third and maybe even a fourth part, the third part is called `funciton minimization`,
it goes like this, you have anonymous function body, and you wanna get cool output without knowing the best output, what a simple minmization will do is, use Math.Random to try different parameters, what more complex minimization will do is will try both parameters, and make steps in the forward direction while amplifying the parameters that work more than the ones that does not, the step size is most important here, as smaller step sizes lead to more precise answers, everything else is handle in the runtime

4. activation, the final problem the current neural networks solve is stock market predictions, or generating situations that satisfy contrains perfectly, imagine a law maker made a maze labirthn of subtractive penalizing restricitve laws for everyone to navigate, with cool activations you can basically find the most efficient path between those restrictions

>>512122302
>I mean I'm just saying we all make choices, so does AI
"""AI""" doesn't make choices. It does the equivalent of following a fixed decision tree.

>these choices aren't "fixed"
But they are fixed, unless you're talking about the PRNG being used to sample from the probability distribution that the model outputs. Which doesn't "choose" anything, either.

>>512122413
>>512122505
So you go through life mindlessly and deterministically following a fixed rule-set that doesn't even exhibit Turing completeness? :^)

>>512121096 (OP)
the globe
dinosaurs
space exploration
covid
the holocaust
911
sandy hook

and the coming fake internet grid blackout blamed on a fake solar flare
and the fake asteroid impact

>>512122595
What if>then causes you to repeat yourself mindlessly like this?
>if I admit im retarded then I lose the argument

>>512123090
Do you go through life mindlessly and deterministically following a fixed rule-set that doesn't even exhibit Turing completeness? Yes or no?

Notice how your retarded algorithm forces you to keep replying even though this simple question permanently stifles your entire shtick. :^)

>>512121757
Technically it does, because every "AI" up to now is still being run on digital hardware, where every transistor is driven either to cutoff (0) or saturation (1), hence if-then-else.
The magic is in the algorithm that combines billions of if-then-else into something meaningful.

>>512121096 (OP)
It is still useful anon

>>512123248
>The magic is in the algorithm that combines billions of if-then-else into something meaningful.
There isn't any "magic algorithm". Your "magical" token predictor does the equivalent of following a decision tree, and I don't mean on the transistor level, I mean in terms of its high-level architecture.

>>512123259
>It is still useful anon
Useful for scamming niggercattle and not much else.

>>512123225
So you never learned?

>>512123446
Do you go through life mindlessly and deterministically following a fixed rule-set that doesn't even exhibit Turing completeness? Yes or no?

Notice how your retarded algorithm forces you to keep replying even though this simple question permanently stifles your entire shtick. :^)

Looking forward to yet another impotent reply of yours, where you concede by default by failing to answer a simple yes/no question.

>>512123515
Crazy how "if touch hot stove then burn myself" is impenetrable logic lol

>>512121960
I will answer since you asked nicely.
Because you admitted yourself that the way you interpret it only works with piecewise linear activation functions.
Even if you steel-man this interpretation and say it applies if you quantize any function (even though you've never shown any understanding of quantization from what I've seen), if you wrote out an if/else statement that represents a layer it would be giant and unintuitive (even for piecewise linear activation functions), as opposed to a diagram of the neurons or even a combined formula representing the layer which would be both more succinct and more representative of the math behind how AI algorithms work.
A lot of things can be reduced to binary logic, in fact anything a computer does, but that doesn't mean it's a good model for understanding what it does.
Here's a good model: AIs produce the most statistically likely output regardless of if it's correct. That's a much more devastating way to say they aren't close to real intelligence, because anyone who believes in AI believes thinking algorithms work on computers already, pointing out that an algorithm uses binary logic is meaningless because everyone knows computers use binary logic.

>>512123663
Do you go through life mindlessly and deterministically following a fixed rule-set that doesn't even exhibit Turing completeness? Yes or no?

Notice how your retarded algorithm forces you to keep replying even though this simple question permanently stifles your entire shtick. :^)

Looking forward to yet another impotent reply of yours, where you concede by default by failing to answer a simple yes/no question.

>>512121096 (OP)
True, as long as it's not sentiment then it's now AI.

>>512123665
>Because you admitted yourself that the way you interpret it only works with piecewise linear activation functions.
What a mentally ill reply. I "admitted" my own argument that automatically refutes you?

>>512123762
I won't argue against low effort replies

>>512123738
The current head of the TKD department has spoken.

>>512123705
Yup, you got no argument for why you're any different than an ai

>>512123898
Good. You're an inbred 70 IQ monkey who shouldn't be talking. Imagine thinking I read any of your slop past your automatic concession.

>>512123427

>>512121294
>n..no u!!

>>512123998
>>Do you go through life mindlessly and deterministically following a fixed rule-set that doesn't even exhibit Turing completeness? Yes or no?
>Yup

>>512123090
He's a summerfag, an underage phoneposter that gets angry when his mother unplugs his game console so he comes here to make threads to get BTFO. I'll accept his concession that he still doesn't understand what AI is though.

>>512124059
Look at every post of yours in this thread. Your are a fleshbot

>>512123998
>>512124164
>>Do you go through life mindlessly and deterministically following a fixed rule-set that doesn't even exhibit Turing completeness? Yes or no?
>Yup

>>512124069
I’ve been stalking him a bit recently. I estimate his age to be 30-35

>I’ve been stalking him a bit recently. I estimate his age to be 30-35
Obsessed tranny. Nothing more rewarding than seeing tards getting mindbroken like that.

>>512124479
There's no way he's 30+. He's about 16. Learning to code like a retard to him is watching youtube videos and then giving up. Look at how tiny his pictures are. I'd be surprised if he even has a PC. This time of year, especially when school is going to start next month the seethe for teenagers is pretty high. Their mother is getting tired of them. They need a release.

>>512124619
I get alerts on my phone whenever you post

>>512124780
>admitting the extent of your obsessive mental illness and asshurt
Ok. Thanks for letting me know.

>>512124942
I simply care about you, Moldova

>>512125062
The asshurt lolcow will reply to this post.

>>512121096 (OP)
I already told you yesterday that you don't need conditional statements to write a functioning AI

>>512125159
I’m just a big fan of your work. :^)

>>512125314
>you don't need conditional statements to write a functioning AI
Explain why this is relevant. You literally can't.

>>512124780
Do you not remember the Vollposter? He was also an underage summerfag, there's the same level of arrogance and stupidity and post count that you can only find in an immature mind that's got lots of testosterone but no actual life experience. Vollposter even learned to lie to say he was 18 simply because he kept getting banned.

Also do you know what happened to Vollposter in the end?

>>512125366
It's relevant because your OP implies AI is a series of conditional clauses in an algorithm while in reality, NN works by summation and applying a nonlinear analytical function as activation

>>512125320
You have to be pretty desperate to try to embarrass me, on an anonymous board, by screen-capping information I deliberately reveal on that very same board.

>>512125511
You've decided not to be anonymous by sticking out like a sore thumb. You make all these threads about how fucking stupid an arrogant you are instead of actually discussing the topic. I'll bet your mother's cunt is still tight though. Where's your dad by the way? Does he know you have regular meltdowns like this?

>>512125729
>mentally ill retard thinks that if he's asshurt about my posts i stop being anonymous
Adding memeflags to the filter. If you want to continue this discussion, let's see which brown country you're from.

>>512121096 (OP)
Why am I seeing your low effort non political threads every day? It's very tiring. There has to be a way for you to get the fuck out

>>512125457
>your OP implies AI is a series of conditional clauses in an algorithm
You have not shown otherwise. Your statement doesn't refute this.

>>512125814
Aww poor baby got triggered. :)

>>512125442
I went through about 1000 of his posts a few days ago in the archives. He does post much more regularly in the summer, so I’ll keep the summerfag theory in mind.

I believe there’s a second Moldovabro who posts in the Ukraine threads, so I’ve had to figure out ways of distinguishing the two. I have most of his reaction images cataloged, and he reuses a lot of the same phrases frequently.

>>512125814
You are essentially the same at a tripfag at this point

>>512125849
>getting brainwashed into an AI-slop-based pseudo-economy is not political
>getting brainwashed into believing that you are just like a token-predicting automaton is not political
>getting your water reserves wasted on AI slop is not political
>etc.
Dumb golems think only the surface-level political circus clown show is political.

>>512125991
You lost an argument. Now I'm living rent-free in your head. This is not my fault.

>>512126009
Personally, I like the Ai revolution of our times. It has made my job much easier. You sound like a butthurt faggot, skipping his asspergers meds.

>>512125865
I have shown you in the last thread
but ok
let W be a mxm matrix with each element representing weight to the other.
let phi be a perelement vector activation function like sigm(X)
let X be the state vector of each node

Then the forward propagation can be calculated by X = sigm(W * X) iterating over your layers. for eg 2 layers you'd have sigm(W * sigm(W * X)) as final state...

You see that's forward propagation without a single IF statement

>>512126339
>I like the Ai revolution of our times.
So suddenly it's the "revolution of our times"? I thought it's not political, definitely not some kind of a psyop targeting dumb goycattle like you. :^)

>>512126399
>I have shown you in the last thread
You have shown nothing in the last thread and you will show nothing in this thread. You don't even know what you need to show, or how to go about it in theory. Let alone trying to actually show it.

>>512125982
How you know he's young as well is he gets triggered by paragraphs. It means his attention span is "cooked" as the kids say. The larp post about being a programmer was to try to add credibility to his arguments (since he has none).

>>512126542
Yep. It's a technological revolution and I can see its tangible benefits. Meanwhile you are here screeching.

>>512127042
>It's a technological revolution and I can see its tangible benefits
Name any "tangible benefits" that are objective and verifiable. You literally won't.

>>512126593
Show me the IF statement in my FP algorithm: >>512126399
You can't. There is none...

>>512127138
I can confidently speak about my job as a researcher. Writing papers and analyzing data has never been easier

>>512127218
Dude, you're 80 IQ and don't understand that your basic premise is true but also irrelevant. You don't even know what you need to show, or how to go about it in theory. Let alone trying to actually show it.

>t-there are no benefits!
To a child like you, sure. But adults use it to get real work done. You should ask them.

>>512127324
Just show the IF statement dude
Its so easy
Where's the "IF...then" line?

>>512121569
>yesterday
>ai makes fusion power

>two more weeks
>ai makes flying cars

>two two more weeks
>ai gets (you) laid

>>512121096 (OP)
LLMs don’t work like this. That’s how old chat bots used to work.

>>512127259
>my job as a researcher. Writing papers and analyzing data has never been easier
You know what? I can sorta believe this. The """science""" establishment has been getting caught repeatedly for publishing nonsensical AI slop. I'm glad being a scammer on the government payroll is now easier for you. Adding your ID to the filter. Piece of trash.

>>512127375
>Where's the "IF...then" line?
Fucking kek. Literally a 70 IQ orc with no abstract thought. You made this mistake ~20 times in a row in the last thread and you just came here to repeat it some more. No more (You)s for you. Imbecile.

>>512127401
>LLMs don’t work like this.
Proof?

>>512127413
Lmao you really are a butthurt nigger. You sound very jealous. It's evident by the emotional pain your posts exude.

>>512127479
You've been proven wrong by counterexample.
Neither FP nor BP algorithms require conditional branches.

Cry and ad hominem all you want - you lost

>>512127218
He said in a past thread the if/else thing is about how you can reduce the output of layers of networks using piecewise linear activation functions to if/else statements. He doesn't care about correctness or broad applicability.

>>512127614
>You've been proven wrong by counterexample.
You are literally retarded. The only relevant counterexample would be to show that your fake """AI""" can do something beyond what decision trees do i.e. prove that it's in a different computational class. Failing that, you could try to show that there is no reasonable way to map one onto the other, despite both being in the same computational calss. I know with 100% certainty that you can't do either of those things. I am also 99% certain you didn't understand this post and will shit out another repetition of "I didn't need to write an if statement to make babby's first ANN".

>>512127550
he's of mixed heritage (German and Russian). so maybe one of his parents lost their meme translator job to AI. AI translation and transcription is really good now.

>>512127859
Notice how your delusional mental illness is causing you to insist that the exact same output is somehow "incorrect" depending on how you implement the same underlying structure.

>>512127896
I'd feel sorry for this poor faggot, if he wasn't so annoying and self righteous

>>512121096 (OP)
You can't prove free will. Humans are if-else-then-end machines. Everything's predetermined and you're only under the illusion that you have choice.

>>512128073
That's correct actually and it has been proven.

>>512128073
>You can't prove free will. Humans are if-else-then-end machines
See what I mean when I say "psyop"? It's a growing horde of these retards thinking humans are essentially biological chatbots.

>>512127980
I hope he keeps seething and writing 100 posts in all these threads. Because then it's only a matter of time before someone figures out who he is and he ruins his life. It will be 100% his own fault too.

>>512128202
Notice how you didn't say that I'm wrong, and that's because you can't. It's over, paisen.

>>512128273
>you didn't say that I'm wrong
I accept your testimony that you are a biobot. There's no discussion to be had with you. I'm just noting that you're the 5th """AI"""-believing mongoloid ITT who claims this.

Your only function here is to demonstrate that this is indeed a psyop.

>>512127974
Correctness of the model you're using to understand AI, not of the output you're using. You can reduce anything a computer can do to if/else statements, but this doesn't help you understand anything about what it does.

>>512128348
Great. Enjoy your psyop, broski! Hope you like it!

>>512127868
>that your fake """AI""" can do something beyond what decision trees do i.e. prove that it's in a different computational class

If riding a car and riding an ant get you from A to B, does that mean ant = car?

>>512128385
You're a dumb, babbling nigger. Please consult >>512127868 for the bottom line of this "discussion".

>>512128448
>didn't understand what he just read award

ATTENTION IS ALL YOU NEED
NIGGER

>>512128496
Addressed and refuted

i fuck yo bitch with weighted perceptrons and pre trained generators. bitch

>>512128552
You sound butthurt but still kekt.

>>512128581
If you actually believe that and aren't just being a spiteful kike, you are 100% mentally ill. No further discussions.

>>512128202
But that's what they are. You don't choose your genes, you don't choose your family and upbringing. Also, the creation of the electrical signal responsible for a conscious thought, precedes you being aware of the thought. And to be frank, your responses are more predictable than those of an AI.
>>512128251
It will be glorious. He will doxx himself, and that's gonna be so fun.

>>512128523
>can't answer again

To you a car is an ant because both move from A to B? Get a load of this retard lmao

>>512128385
>You can reduce anything a computer can do to if/else statements
you still need the concept of iteration (like a loop or recursive function call). the transformers (key part of LLMs) actually require this in multiple places, even bare bones neural networks are represented as matrices which are multiplied which requires iteration. so even at a basic level he doesn't understand what it is he's talking about.

to steelman him, we can say he thinks that LLMs are "narrow AI" much like the old expert systems (which domain specific knowledge was encoded into if else statements), and that LLMs are just another example of this. but the problem is that if else statements only produce results defined in the rules. he doesn't understand that LLMs are showing some (shallow) generalization capability. that's what's interesting and useful about them. all he has is cope and trying to insist it doesn't meet his personal standards, which nobody cares about.

>>512128837
Get your mental illness treated.

>>512128251
I probed him with this: >>512125320

Interesting reaction.

I’ll share more when I have it. In the meantime, I have an alert system set up so that I can converse with him whenever he’s on (with an international VPN as needed). I have a few potential leads, but as I said there are multiple people from Moldova so I must use fuzzy logic — like LLMs.

>>512128860
You implied equivalence due to similarity
So is a car an ant underneath? Yes or no? both are similar transportation classes

>>512128960
See >>512128860

>>512128937
> I must use fuzzy logic — like LLMs.
Kekd and checkd

>>512128999
>can't answer
>ad hominem

you lost again
Imagine being such a loser lmao

>"""AI""" niggers getting so mad about my calling their """AI""" god that they're threatening to doxx me and ruin my life

>>512129076
See >>512127868

I still think the world's biggest ever rug pull is coming for bitcoin.

>>512128937
he's such a newfag doesn't even understand how the board works.

YmVzdCB3YXkgdG8gZXhsaWNpdCBpbmZvIGlzIHRvIGluc3VsdCBoaW0=

SGUgdGhpbmtzIHNlbGYtYnVtcGluZyBpcyBhIHRoaW5nLg==

>>512129144
>I still think the world's biggest ever rug pull is coming for bitcoin.
Definitely another massive psyop but the idea isn't that dumb in principle. The problem is that niggercattle only cares about using it for speculation.

>>512129140
loser
If you claaim something, the burden of proof is upon you, not me
Proof EVERY NN consists of conditional branches
go ahead

>>512129260
>If you claaim something
You did. Dozens of times. See >>512127868
then see >>512128860

>>512129300
You haven't provided proof to your OP.
Go on, we'll wait
>inb4 doesn't deliver

>>512129240
I think there is a point where prices will drop and people with huge amounts of bitcoin will want to cash out and it will precipitate an enormous drop, causing more to sell and so on.

>>512129370
So you back down from all the claims you made and it all dissolves to "you haven't provided proof"?

>>512129391
>I think there is a point where prices will drop and people with huge amounts of bitcoin will want to cash out and it will precipitate an enormous drop, causing more to sell and so on.
Probably. When """governments""" start banning exchanges and these niggers fear they can no longer convert their goycoins into goybux.

>>512121096 (OP)
can ai generate true random

>>512129260
He claims he doesn't believe AI exists yet he makes the same thread every day about it and seethes for 100 posts:
https://archive.4plebs.org/pol/thread/511225153/#511225833

>>512129449
No my claim is still truthful.
But as I said, doesn't deliver again
really low iq energy
yesterday you didn't even know what an analytical function was, you thought the if clause was in the activation XDDD
don't you have highschool to attend?

>>512129599
>my claim
The one you don't know how to prove?

>>512129772
That's not how logic works. If you claim something in your OP, the burden of proof lies on you, not on the people telling you you're wrong.

>>512129982
>If you claim something, the burden of proof lies on you
So are you going to prove your claims?

>That's not how logic works.
I know you're 70 IQ and unable to comprehend any ML-related topics, but at least learn from this thread that """burden of proof""" has nothing to do with "how logic works".

>>512129105
As I said earlier, I just care about you. You’re not alone in this world. I’ll always be there.
>4u

>>512130061
You're legitimately the first anon I met that has lost at pol.

>>512130256
What were you trying to prove by drawing a circle on a picture? Are you actually 8 years old?

>>512130314
try reading what it says
in case you have eye damage

Oh he is just a teenager. 8 years old is an insult to him. Are you looking forward to going back to school next month kid?

>>512130412
I don't understand. You were claiming something about "how logic works". Which system of logic is the image supposed to be describing? Once upon a time I studied type systems, so I'm familiar with quite a few. Name the one whose working involve "burden of proof". :^)

>>512130548
>can't prove his OP claim
>derails into irrelevant shit
You had ONE sticky to read.
Just ONE

>>512130423
why does he uspet you shillniggers so much?
sounds like hes right over target

>Once upon a time
That is how fairytales start.
>I studied type systems
Yet you fail to grasp the absolute basics of programming. You clearly didn't study them very well.
Where is his father? Does he know he's on a website that's 18+?

>>512130854
See >>512130548
By the way, why do you keep making claims and then failing to support them all while screeching about some jewish """burden of proof""" that others allegedly have?

>>512130924
A bunch of them have literally admitted ITT that they think they're essentially the same as chatbots. They are getting assblasted because by making fun of chatbots, I am making fun of what they consider to be their true nature. Simple as.

>>512121569
midwit

> about some jewish """burden of proof""
>jewish

Why are you using this board if you disagree with the only sticky here generally agreed upon?
>>503281217
Go back

>>512131214
If you agree with the sticky, why do you keep making claims and then refusing to support them? Is basic coherence not a part of "how logic works" for you?

>>512130924
Oh great we have a lightbulb thinker who thinks that there must be something to OP's retardation because others are correcting him.

Does it not occur to that something like LLMs or even just NNs can be useful despite things being in a massive bubble? Or does your brainlet mind go: "it all has to be a scam" because you couldn't be bothered to learn about it? Jewish usury, money counterfeiting cause these boom and bust cycles in the first place. The exact same thing happened more than 20 years ago.

You are the brainlet boomer that is saying "websites are scams" because you don't know what they are and you see a Jewish credit bubble and it feels wrong to you. You have no ability to think critically, hence resorting to feels and lightbulbing.

>>512131059
Thought you filtered me? Still upset I see. Good. You should be, school is starting soon.

>>512131258
Why do you refuse to proof your OP?
And instead rage at the sticky like a newfag

Imagine being forced to drive one of those electric cuck carts, and then it crashes into a lamp post because the driving system was vibe-coded by some brown """AI""" believer who takes code snippets from "reasoning" models and puts them into production codebase.

>>512131422
>Why do you refuse to proof your OP?
I actually explained to you why the pic makes sense yesterday but you're 70 IQ and couldn't grasp it.

>>512121932
>be me
>google how to generate a UPC-13+5 barcode in python
>AI assistant gives me some code snippet which I instantly know isn't going to work because it references a library that I just read all the documentation for
>copy and paste the code snippet just to see it crash
>"people" like you still consider this god-like super intelligence

>>512131526
>I actually explained to you why the pic makes sense yesterday
link to this post archive
>inb4 can't find hurr

>>512131329
A 1pbtid German swooped in to the rescue in some other threads of Moldovanon. It’s a pattern my biobot brain noticed. Could be nothing, who is to say?

>>512121096 (OP)
Q and flat earth are good contenders.

>>512131593
Last time we were specifically discussing deep ANNs with ReLU activation functions. There every layer is equivalent to a piecewise-linear function, which is equivalent to some conditionals for selecting the appropriate linear piece. That's a pretty popular architecture for various kinds of """AI""" and the classic transformer also features such ANNs. You can have fancier activation functions, but then you can make the same argument in terms of max-affine spline operators - every layer is still a piecewise function. Seethe about it.

>>512131710
When he really wants to reply but can't he'll call for help on discord to do it for him it seems.

Also what kind of person makes the same thread every day and writes 60-100 posts about something they don't think exists? Did he ever actually explain why or is he just an Eliza chatbot doing his if else with canned responses?

>>512131710
very shill bot post you make there
everything you shills post in this thread makes my spider senses tingle

>>512131791
>flat earth
At least the ground looks reasonably flat. It's dumb but not as dumb as thinking ChatGPT is intelligent despite the evidence of your own eyes.

>>512132066
no link lmao
>Last time we were specifically discussing deep ANNs with ReLU activation functions
No we were not, you must mistake me for another anon

>There every layer is equivalent to a piecewise-linear function,
a layer isn't activation lmao.

>You can have fancier activation functions, but then you can make the same argument in terms of max-affine spline operators
You can also have simple activation such as a sigmoid which is entirely analytical.

But since you posted no link anyway, i guess you conceded...

>>512132423
Mentally ill mongoloid. Go dig up the thread yourself, or ask one of your obsessed tranny buddies. I don't care. The justification for the pic is provided here, like it or not: >>512132066

>>512121096 (OP)
>Name a dumber psyop.

Islam

>>512128073
humans work completely different
the simple obvious proof is that "artificial AI" needs need huge amounts of data
a human doesnt need that
a humand understands, the machine doesnt
a human can learn to drive a car after a few hours, Tesla still cant do it, with billions of dollars

>>512132542
>Go dig up the thread yourself,
>can't post anything
>still can't proof his OP even
>assmad and malding into ad hominem

kek you lost

>>512132288
I’m shillin’ like a villain, G*rmany. I like your typing style!

>>512132066
>There every layer is equivalent to a piecewise-linear function
Wrong.

If-else = runtime branching, discrete jumps, non-differentiable.
NN layers = fixed algebraic graph, continuous transforms, differentiable (almost) everywhere.

Do you just enjoy being wrong or something?

>>512132288
Thank you for sharing your feelings. It's all about the vibe isn't it? It's good to stick up for your friends too.

>>512132670
>the simple obvious proof is that "artificial AI" needs need huge amounts of data
Humans also need "data", but the machine needs specifically data collected, processed and contextualized by humans.

>>512132682
You can keep lying and seething but anyone can read >>512132066 for himself for a justification of the pic.

>>512132738
>Humans also need "data"
not anywhere the same amounts
because humans understand
the machines only work with statistical models, which means endless amounts of data
as I said, human understand what hes doing when he drives car, the machine doesnt

>>512132413
Yes the "ai" cult that is currently forming is definitely retarded but I do understand why so many people fall for it. It's quite overwhelming and magical to people who don't understand what is going on under the hood. I've had enough trouble convincing certain people that Wikipedia and Google are captured and seriously biased.

>>512132799
1) Where's the link?
2) Since no link: You have to prove where the conditional logic lies, so far you haven't.

daily reminder OP is a pathetic subhuman ape with no life spamming the same thread over and over, lmao
https://archive.4plebs.org/pol/search/image/29M733yMtcKG5GQXMrEYrQ/

>>512132932
>not anywhere the same amounts
Of course. Brains don't learn by fucking gradient descent.

>>512132987
and another AI shill shows up

>>512121096 (OP)
It TELLS them they're right. I mean, it's designed for lonely incels who think they're smart and have never had someone headbob at them.
Good stuff. Just a pain we have to endure them posting their "genius" conversations and equally midwit ceo's trying to "leverage" it for their companies.

As someone else on her said, great as a personal assistant, garbage researcher/worker.

>>512132987
>another retard admits to thinking he's essentially like a chatbot
6th case just in this thread. How is this not a psyop, again? Imagine living in a society where this cancerous opinion has trickled all the way down to the average normie.

Damage a surveillance drone? That's aggravated battery, that bot is just like people! Now you're going to get assraped in prison for 15 years.

>>512132984
For reference in future threads: >>512132712 he doesn't actually understand either of the terms he's using and refused to engage. He can only seethe.

>>512132984
You can keep lying and seething but anyone can read >>512132066 for himself for a justification of the pic.

>b-b-but where's the heckin' link for the last time I got BTFO
Completely irrelevant. If you care, go dig it up.

>>512133125
>>512133217
subhuman ape schizos are like a broken LLM, lmao

>>512133327
So to sum it up, you think the conditional branching lies in the piecewise reLU definition? Is that correct?

>>512133408
no, he can drive a car, your AI machines cant

>>512132976
>It's quite overwhelming and magical to people who don't understand what is going on under the hood. I've had enough trouble convincing certain people that Wikipedia and Google are captured and seriously biased.
I guess, but if you play around with it enough and you're >90 IQ, eventually you'll do something that makes the seams show and the illusion breaks. No wonder they want to produce a retarded brown subrace. Convergence between man and machine will happen not through the ascent of the machine but through the descent of man.

>>512133433
>you think the conditional branching lies in the piecewise reLU definition
I don't know what this retarded drivel means. My post explains what I think very specifically.

>>512133517
lol

>>512121096 (OP)
Epstein is the dumbest psyop. Some coomer pedo offs himself, and everyone thinks the fucker is the Rosetta Stone that will explain all the geopolitical relationships they're too fucking stupid to understand otherwise.

>>512133183
The one by (((OpenAI))) has been specifically finetuned to be agreeable to the user because it makes normies interact with it more and feed them more data. That is very Jewish.

>>512133517
Useful information, thank you. Keep sticking up for your boyfriend like a good NPC, it's not obvious at all.

>>512125814
Relax, guy. One day a missile will strike you in the head from Russia while you sleep. Then none of this will have mattered.

>>512133697
>My post explains what I think very specifically.
Then surely you'll have no problem telling me again where the conditional branching occurs. You said in the reLU activation right? Yes or no?

>>512133408
Reading and kek'ing. I should start reading my own posts in the archive. By far the best thing that happened to this board.

>>512133874
As soon as you quote something where I claimed anything about "conditional branching occurring".

>>512133712
how many billion more miles does the machine need before it figures it out
a human can do it in two days

>>512134044
>As soon as you quote something where I claimed anything about "conditional branching occurring".
Don't nit-pick. You wrote:
>which is equivalent to some conditionals for selecting the appropriate linear piece.

So the ReLU activation function is where the "conditionals" occure, according to your argument? Correct or no?

>>512121096 (OP)
the only people talking about AI "not being real intelligence" are professors who try to get people to believe thinking is a magical process only they possess the "real keys to doing" or those who got indoctrinated by those professors.

>>512133908
>Reading and kek'ing. I should start reading my own posts in the archive. By far the best thing that happened to this board.
lmao
you're a subhuman moldovan schizo retard spammer but in your own mind you're some kind of genius
the pysop is your own life, lmao

>>512134321
It's a plinko machine man, come'on.

>>512134321
another shills claiming hes a chatbot
maybe he is

>>512134246
>You wrote:
>>which is equivalent to
Yes. Equivalent to. So why are you asking me retarded questions about claims I didn't make? I can write a literal decision tree and then rewrite it to run on the GPU without conditional branching occurring anywhere. Doesn't change the nature of the thing.

>>512134321
Example #7 of a brainwashed golem thinking he's essentially a biological chatbot. Imagine living in a society where this cancerous opinion has trickled all the way down to the average normie.

Damage a surveillance drone? That's aggravated battery, that bot is just like people! Now you're going to get assraped in prison for 15 years.

>>512134473
>I can write a literal decision tree and then rewrite it to run on the GPU without conditional branching
So do that. You can't though. You're just a nocoder larper.

>>512134473
Ok so pick a different activation.
Pick the logistics function.
That one has no "conditionals"
Also whatever you write to approximate it will have digitalisation error ok? So technically it's not equivalent to the analytical, so for now no GPU, only piece of paper.

Where's the "conditionals" in y(x)=1/(1+exp(-x))?

>>512134560
>74 pbtid lol
Yes, this slope headed inbred slav is actually calling other people here obsessed
Post forehead lmao

>>512134803
>Ok so pick a different activation.
See >>512132066

>>512134835
We're already past that. Replace ReLU with sigm. What now? Where's the "conditional"?

>>512134924
See >>512132066
>You can have fancier activation functions, but then you can make the same argument in terms of max-affine spline operators - every layer is still a piecewise function. Seethe about it.

>>512134332
You just know that even when his mother unplugs the router at night at bedtime he lies awake still thinking about these threads and seething. /pol/ will break him. They all break. It just takes time. Might be a year or two from now, but they all come here thinking they can change this place and it ends up changing them.

>>512134981
>every layer is still a piecewise function
But y(x)=1/(1+exp(-x)) isn't piecewise.
It's analytical and continuous, which means continuously differentiable along entire R

>>512123392
From a high-level perspective, LLMs consist of MLPs and attention. If you go a bit lower, it's basically all matrix multiplication. If you want to claim it to be equivalent to a decision tree, you have to go very low level.

>>512123515
Your memory isn't infinite, so you are not truly turing complete senpai.

>>512135104
Mentally ill retard.

>>512135153
Hurts

>>512135126
>If you want to claim it to be equivalent to a decision tree, you have to go very low level.
No, I really don't. See >>512132066.

>>512135126
>Your memory isn't infinite, so you are not truly turing complete senpai.
Kek. What a hill to die on.

>>512135153
What don't you agree with?
Here is the proof it's analytical
https://en.wikipedia.org/wiki/Logistic_function#Taylor_series

>>512134560
they will make you starve to feed their machine lunacy

>>512133553
The Coudenhove-Kalergi plan is a thing and there are still elites who envision a future with superhuman elites and goblin-like underclass like Yuval Noah Harari. The problem is that they are only good for manual labor. I think mixing high IQ and low IQ ethnic groups is a divide and conquer tactic. The point is to make uprisings like the Germans did impossible but not to completely crush human ingenuity. That's how I would do it anyway.

>>512135320
I agree with everything except for the part where your mental illness causes you to think your statements refute something I said.

>>512135413
>your statements refute something I said.
Where does it support what you claim tho?
Where do you need a "conditional" to calculate the logistics function?

>>512121886
What is your brain if not a big list of if/elses

>80+ posts
Holy shit, get a life.

>>512135406
these are the people that shill AI

>>512135320
I don't think he actually understands what analytical means, rather than him actually admit that he tosses out insults to deflect from his ignorance. If he refused to engage on two short sentences explaining to him the differences between both then there's no hope he will read a wikipedia page.

>>512135409
>The point is to make uprisings like the Germans did impossible
That is right on the money. Germans were intelligent, diligent and worked as a collective ethnic group and saw through what the Jew was doing. If they produce some kind of mullatto type "race" that has no ethos and is in the 80-90IQ range, that's never going to ever reach a point where they challenge the Jews again.

>>512135595
Where did I claim you need a conditional to calculate a logistics function? You legit have some kind of mental retardation, anon. Unironically take a break.

>>512135893
logistic*

>>512135874
answer this :>>512134233

>>512121096 (OP)
covid was a huge one, waiting for neural networks in the psyop graveyard

>>512135893
>Where did I claim you need a conditional to calculate a logistics function?

So do one need them in this example, or not?
Make up your mind.

>>512135409
They'll just use genetic modification to maintain a small cadre of slave savants. Producing geniuses the traditional way requires keeping too many "useless eaters" fed per genius.

>>512136051
Are we back to "but I don't heckin' need conditionals to do X" again?

>>512136210
Answer the question!
To calculate a logistic function, do one necessary need a "conditional"?

>>512136288
>To calculate a logistic function, do one necessary need a "conditional"?
Nope. And what of it?

What exactly has artificial stupidity accomplished except creating slop images that boomers on fb confuse with reality and creating tons and tons of shortform crap on jewtube?

>>512121096 (OP)
this just in AI is a search engine - been around for 30 fucking years

>>512136329
Ok great.
so we don't need "conditionals" to calculate a logistics function. I agree Next, do we need "conditionals" for
A) Summation
B) Multiplication
?

>>512136329
>and what of it
lmao, great thread

>>512136389
>What exactly has artificial stupidity accomplished except creating slop images that boomers on fb confuse with reality and creating tons and tons of shortform crap on jewtube?
Literally convincing a growing segment of the population that minds aren't real and that humans are just bio-bots.

>>512136491
>I don't need conditionals to do X
Yes, nigger, you've been screeching this for days. I'm just waiting for the logic that brings you from this observation to some kind of argument against the pic. Actual logic. Not a straight leap to your incorrect conclusion.

That logic won't come.

>>512136389
its actually quite impresive in certain areas

https://deepmind.google/discover/blog/alphaevolve-a-gemini-powered-coding-agent-for-designing-advanced-algorithms/

>AlphaEvolve’s procedure found an algorithm to multiply 4x4 complex-valued matrices using 48 scalar multiplications, improving upon Strassen’s 1969 algorithm that was previously known as the best in this setting. This finding demonstrates a significant advance over our previous work, AlphaTensor, which specialized in matrix multiplication algorithms, and for 4x4 matrices, only found improvements for binary arithmetic.

>>512136621
Please answer the questions first!
do we need "conditionals" for
A) Summation
B) Multiplication
?

>>512135949
I can explain that no problem. Self driving was showing promise, but there are plenty of "edge cases", like where you have bicycle on the back of a car that the models get confused and assume a bicycle is floating and moving across the road. The "fix" was a hack, where Pajeets at Tesla decide yeah let's invent a new type of vehicle type where it's car+bicycle to get around the problem. Rather than taking a step back and actually fixing it properly.

Guys like Elon (who hire Pajeets in the first place to replace competent white engineers who tell him he's wrong) are outright scam artists, who believe in "fake it until you make it" as a motto - which is why he's been saying "full self driving" is coming next year for like 5 years now. That doesn't mean that we can never get full self driving. It means that it's just an unsolved problem. As a young person, you should be excited to work on something like that. It's easy to be a cynic. In fact, any retard can be a cynic.

There will be a moment, similar to what happened with LLMs where, you will see more generalising and emergent behaviour. Can I tell you when that is going to be? Next year? 5 years from now? I can't I don't actually work in this field. I can tell you that if you're just believing guys like Musk you'll only be disappointed though. He's not an engineer, he can't even code. His job is to go up on stage and hype things up so people buy stocks and he can get more credit. You will have to learn to seperate characters like this (and like Sam Altman) from the technology itself.

>>512136511
Yeah I remember that lmfao, they claimed consciousness isn't biological because their chatbot uses emojis and because some schizo engineer said his chatbot broke containment

>>512123898
>>512124023
hey guys, just posting to let you all know i thoroughly enjoyed your little interaction there. i love how the leaf started out all agressive and shit insulting the moldavian for no reason, but 2 posts later you guys switched roles and the village rapist became super offensive out of the blue and the leaf didn't even respond lol very unsual interaction don't usually see that kind of stuff over here. just my thoughts on the whole thing lol peace out

>>512136682
You DUMB motherfucker. The question isn't whether or not you can do something without conditional. The question is whether or not there is something you can do with your regression activation function, that can't be just as easily done using piecewise functions. And the answer is 'no'.

>>512136648
Yes it's a very good calculator, exactly like how all computers crunch numbers better than humans do, what's the point?

>>512136770
>There will be a moment, similar to what happened with LLMs where, you will see more generalising and emergent behaviour.
2 billion more miles

>>512136936
first answer the question!
do we need "conditionals" for
A) Summation
B) Multiplication
?

>>512137067
Nope. No conditionals needed for any of that. What's the next step of your master plan? I know what it is. Leaping to an incorrect conclusion.

>>512136968
You seemed to miss the part where I said they were taking the wrong approach? I didn't actually say they need more data either. What is this cope? You can either learn about the technology or seethe in ignorance. You can't be German, Germans love discussion technology. You're probably some fucking retarded Arab or Turk. You have to go back.

>>512136968
2 more layers. 2 more data centers. 2 more petabytes of data. 2 more yottawatts of power. 2 more decades. 2 more trillions in investor money. Intelligent machines are coming. :^)

>>512137167
Ok great! So to sum it up, there's no necessary "conditionals" in:
A) Summation
B) Multiplication
C) applying the ligistic function sigm(x)

Correct so far?

>>512137318
No conditionals needed. We've covered this multiple times. I like how you just keep stalling because there is no logical next step for you. :^)

>>512137548
Because he’s won the debate.

>>512137548
good.
Now in equation
>sigm(W * sigm(W * X))
where W is mxm matrix and X is m vector, do you see any other operations besides A) B) and C)?

>>512136151
>Producing geniuses the traditional way requires keeping too many "useless eaters" fed per genius.
Yes I was being too optimistic. Producing geniuses capable of true innovation requires a large high quality population and specific conditions. We have already lost the ability to produce enough of them.

>>512137727
Nope. Just make the rest of your "argument" in a single post instead of stalling. I just want to get to the retarded nonsequitur that you inevitably run into.

>>512137176
2 more approaches

>>512137859

>>512137176
im not against progress
but im also not stupid to think humanity can recreate science fiction from Star Trek because someone throws bllions at it
thats not how it works

>>512137861
>Nope.

Good. Because that's literally it. That's the whole AI forward propagation algorithm (FP). It only consists of A) B) and C) operations. In principle we're done - X is your input, the result is NNs output. That's the simplest case and at no point did we employ "conditionals". If you feel fancy pick up a pen and calculate it yourself on a piece of paper, no PC needed. QED

>>512121096 (OP)
you don't know me.

>>512138199
Notice how your post makes no attempt to go from babby's first NN explanation, to proving that the pic is wrong. None at all. It's just an arbitrary leap.

hello
I'm here to nitpick over the definition of 'conditional' also
please to stroke?

>>512138242
Is this real?

>>512138383
Here's a (You) because you posted Alina. I'm still waiting for anyone to explain what a sigmoid function can make a neural net do that a piecewise function can't do just as easily.

>>512138292
>to proving that the pic is wrong
Proof by counterexample. We just concluded designing a NN with no need for "conditionals" whatsoever

>>512138563
That's great, but I never said you need conditionals for a NN. I said your conditional-free NN is still equivalent to a decision tree. At no point did you approach refuting this.

I'm still waiting for anyone to explain what a sigmoid function can make a neural net do that a piecewise function can't do just as easily.

>>512138527
I like sigmoids because they have smooth curves

>>512138712
As you can see, Alina has smooth curves but she also has boundary conditions.

>>512138667
>That's great, but I never said you need conditionals for a NN.
Your OP implies behind every AI is "conditionals". Just designing one disproves it which is what we did.

>I'm still waiting for anyone to explain what a sigmoid function can make a neural net do that a piecewise function can't do just as easily.

Differentiation over entire R.
He said it in >>512132712

>>512138194
And who in this thread says that we need to be spending billions on this? Especially public money? I actually don't care if some billionaire wants to put up their *own* personal money on it. But if you look at something like Tesla, they have a history of overpromising and underdelivering, and they rely on gullible members of the public to buy stock to prop it up. I don't agree with that. And if you look at his other companies it's tied to government contracts (X recently got a contract with the pentagon because they dazzled then with "AI"). I don't agree with *any* of those things.

>recreate science fiction from Star Trek because
You're too young to know this but for the longest time there were many things in the old Star Trek, like the communicator or being able to talk to the computer that seemed like we'd never get there. When we have them today. That's not to say that the march of progress is guaranteed, it depends on whether intelligent people decide to work on them. No amount of Pajeets, not even billions of them, will ever invent anything like a transformer. It's all the white man, and Jews and Pajeets are only good at scamming and taking credit for other people's work.

>>512138891
>Your OP implies behind every AI is "conditionals"
Please consult >>512132066 again.

>>512138981
We already past that
you agreed there's no conditionals in A B and C, remember? So ypur statement is incorrect - just because you can approximate a sigmoid with a PWL function, doesn't mean a sigmoid IS a PWL function.

Is it so difficult to understand one is analytical while the other isn't?

>>512139185
>approximate a sigmoid with a PWL function
I can approximate it with all kinds of piecewise functions, not just linear ones. So what can your sigmoid-based NN do that a composition of piecewise-defined functions can't do just as easily?

>>512139283
> So what can your sigmoid-based NN do that a composition of piecewise-defined functions can't do just as easily?

1) The fact you agreed conceptually there's a difference between analytical activation and PW(L) activation proves there's two classes of NN already. But in case you're interested

2) Small-signal operation along the discontinuity. One case will be a linear amplifier, the other a rectifier. Fundamentally different operation mode

>>512139601
So what can your sigmoid-based NN do that a composition of piecewise-defined functions can't do just as easily?

Notice how you will deflect this question for the 5th time. Because the answer is nothing. I can take your ANN and boil it down to a decision tree with piecewise-defined functions and have it produce the same outputs.

It is no more powerful than a decision tree. It is also easily translatable into one. In fact, doing so is a good way to analyze and understand what your gay blackbox is actually doing. Seethe about it.

>>512138972
my point was that these machiens dont work like humans do, they are completely different
its only shills claiming otherwise
Tesla isnt the only one working at self driving cars
and all they achieved is geo fenced mobliles in specific areas
the real world is much more complex than these machines are able to comprehened
a human can do it after two days

>>512138863
an important observation, but the boundaries are literally formed from her curves
they don't actually exist as an independent 'thing'

>>512139816
>Notice how you will deflect this question

see >>512139601 2)

If you operate a small signal along the ReLU discontinuity, you get distortion. You don't get distortion using a sigmoid operating at ANY point, because it's analytical. There's your difference

>>512139816
You can turn a NN into a huge decision tree on paper, but that doesn’t mean they’re the same thing. Neural nets learn and generalize by adjusting continuous weights, not by splitting data like a tree.

>>512140009
>If you operate a small signal along the ReLU discontinuity
Who's talking about ReLUs now? I like how every single post of yours shows symptoms of mental illness and mental retardation.

>>512139884
Who is claiming they work the same? Can you even quote anyone in this thread for these claims you keep pulling out of your ass? Artificial neural networks have borrowed some concepts from real brains but yes, they're not "the same", they learn differently and are even invoked differently than their analog counterparts. What is even your point here? Just make shit up to argue about when you can't admit that you're wrong?

>>512121569
May we see the "photorealistic" images?

>>512140011
>same computational class
>easy to translate one into the other and get the same outputs
>such translation actually gives you a more useful representation suitable for analysis
Seethe about it.

>>512139927
Now you're just getting autistic.

>>512140309
You.
You linked to >>512132066 like a dozen times.
Your confusion stens from the fact you can't tell what a Taylor series is and what analytical means. Don't worry in uni they'll teach you in maths freshman year. And once you work through fourier transforms you'll see what functions cause distortion, what cause rectification and what dont. best of luck buddy, but your OP is refuted anyway as approximation=/= equivalence

>>512140531
Your translation loses what makes the model powerful: gradient-based learning, abstraction, and efficient generalization.

>>512140714
You are legit mentally ill, 100%, no further discussion.

>>512140724
Nonsensical word salad from a low IQ desperately trying to save face after getting obliterated.

>>512140790
approximation=/= equivalence will be written on your meme’s gravestone

Seethe. Russia smoked your ass

>>512140423
He just proved that you are a child of Greenspan-Bernanke.

>>512140667
thanks
if we're gonna get to the bottom of this infinite generative thing, we'll have to dive deep

>>512140947
Since you can't show a single thing one can do that the other can't, they're in the same computational class. There is no discussion to be had about this. Your mentally ill seething doesn't change reality.

>>512140724
I'm just convinced he doesn't understand derivatives at this point. I mean you can't be so blatant as to assume it's equal...

>>512140877
>>512141001
Cool insults. Still doesn’t change the fact that practical behavior, learning dynamics, and generalization matter more than abstract representational equivalence. You’re stuck on “can be represented as,” while the rest of us are talking about what actually works and why. Keep coping.

…300

>>512140998
>infinite generative thing
This exists only in your head. In reality it's just a fixed series of steps always producing the same outputs, strictly less powerful than a Conway's Game of Life.

>>512141088
>practical behavior, learning dynamics, and generalization matter
Same inputs, same outputs. "Learning dynamics", "increasing awareness" or "optimizing workflows" doesn't figure into it.

>>512141201
Same inputs, same outputs, different systems.

A lookup table gives the same outputs too. What matters is how the system gets there: learning, generalizing, adapting. That’s what you keep ignoring because your argument can’t handle it.

I’ll be seeing you :^)

>106 posts by this ID
whew lad. you think you're cooking but you got absolutely cooked again by the Russian. What a melty you're having.

>>512141606
>what matters is spreading awareness, maintaining automated capabilities and promoting inclusivity
Nope. What matters is same computational class, same outputs for same inputs and easy-to-see structural isomorphism. You lost. Move on.

>>512141643
I'll be living rent-free in your head for months to come.

>>512141868
Computational class and output matching say nothing about why a model is useful.

By your logic, all software is the same because it’s Turing-equivalent.

>>512142055
Same computational class. Same output for the same input. Blatantly apparent isomorphism. You lost. Move on.

>>512141923
Rent? What would you know about that? It's obvious you still live with your mum lol.

>>512142055
>By your logic, all software is the same because it’s Turing-equivalent.
And this is obviously false. The Game of Life program I wrote as a 10 years old is more powerful than the transformer architecture.

>>512142113
A hash map is in the same computational class and gives same output for same input, too.

Doesn’t mean it learns, generalizes, or adapts.

You’re confusing representational equivalence with functional behavior. That’s a category mistake.

>>512142151
You'll be lashing out at random Moldova flags for months after I'm gone. I scarred you. Every time some literal who shows up and starts seething for no reason, it brings me a bit of satisfaction to know I did literal damage to your mind.

>>512121569
It can't create original concepts like a human intelligence can, it only recombines and rehashes existing material. It's an input:output machine with no unique thoughts because it cannot think; it doesn't emulate the human brain. Because of this, it often repeats things from incorrect sources and can even contradict itself in conversations. Calling this algorithm an "intelligence" is the most retarded misnomer.

>>512142260
>A hash map is in the same computational class
Proof?

>B-b-but does it poopoo, peepee?
Same computational class. Same output for the same input. Blatantly apparent isomorphism. You lost. Move on.

>>512142321
No one’s disputing equivalence on paper. we’re pointing out that learning systems behave differently than static mappings. If you can’t engage with that, you’re not actually arguing about AI. You’re just flexing a technicality like a faggot

>>512142266
>Every time some literal who shows up and starts seething for no reason
That's you though. You're a lolcow here for my amusement.

>>512142492
>No one’s disputing equivalence on paper.
Your mentally ill cult shat out hundreds of posts trying to disprove just that equivalence. Now it will shit out hundreds of posts trying to save face and move the goalpost.

> learning systems behave differently than static mappings
Same input -> same output. Same behavior. Both are static. You are mentally ill.

>>512142565
You'll be lashing out at random Moldova flags for months after I'm gone. I scarred you. Every time some literal who shows up and starts seething for no reason, it brings me a bit of satisfaction to know I did literal damage to your mind.

>>512142626
You’ve made it clear you can’t or won’t distinguish between what a system computes and how it learns to compute it.

>>512142701
Your "system" doesn't "learn" anything. It's literally a static function just like the decision tree.

>>512142658
There's nothing more NPC-like than spamming a pasta repeatedly because you're upset.

>>512142313
It's just marketing. It's dubbed as "artificial intelligence" because it sounds flashy and makes it sell better, not because it's some psyop to control le normgroid sheeple or something. Retards will drop more money on something if you can make it sound nore sophisticated.

>>512121217
Isn't in the other way? Normies are pretty happy with this new kind of faggotry

>>512142761
Then train it again and get a different model. If it’s just a static function, explain why the function changes after training. You can’t.

>>512142800
You'll be lashing out at random Moldova flags for months after I'm gone. I scarred you. Every time some literal who shows up and starts seething for no reason, it brings me a bit of satisfaction to know I did literal damage to your mind.

Notice how you can't stop addressing me or otherwise talking about me.

>>512142819
>Then train it again and get a different model.
Then train "it" again and get a different decision tree.

>>512142819
>explain why the function changes after training.
It doesn't. You are actually having a psychotic episode right now and thinking your static weights are magically changing as you run tthe model.

>>512142910
>You are actually having a psychotic episode right now
Bro you're at 117 posts. It might be time to calm down and take the Ls.

ITT: 8 separate """AI"""-believing retards admit they think they are biological chatbots.

>>512142870
tree doesn’t learn once it’s built, NNs learn during optimization

>>512142999
>tree doesn’t learn once it’s built
A NN doesn't learn once it's built.

>>512142981
If AI isn't real, how come you've been seething for 119 posts about it? Do you also do this with other things like santa claus and the tooth fairy?