Philosophy of Science General: AI Winter

[Anonymous No.14398452]

>>14395650
>AI winter was caused by the fact that hardware wasn't performant enough and data storage was insufficient to store the vast amounts of data required to properly train various learning algorithms. Considering how cheap multi terabyte storage is now and how we have machines many times more powerful, not to mention gpus that can be used for general purpose problems like AI we aren't heading towards any AI winter.

Deep learning algorithms like CNN, RNN, and so forth are all significantly limited in their applications. Yes, there will be more automation in the upcoming years, leading to large loss in many jobs, but there won't be autonomous artificial general intelligence. AI Winters are caused by excessive hype, from figures like Minsky and Musk, and the future failing to live up to such hype. That's not to say these deep learning algorithms do not have any future applications such as self-driving cars, better face recognition, better game AI (e.g., AlphaGo), and much more, but it is to point out that they won't lead to the fantastical sci-fi vision you have in mind.

"My own Singularity prediction: I predict that in 2030, futurists will be forecasting the Singularity for 2060-2065."
-- François Chollet (creator of Keras)

[Anonymous No.14398586]

so it might get bumped back a couple decades, still happening

[Anonymous No.14398595]

>>14398586
It won't happen, you dumb progressivist faggot. There will be no autonomous artificial general intelligence, advanced brain-computer interfaces, space colonization, or any of that sci-fi bullshit. Honestly, all transhumanists and naive scientific progressivists, like you, deserve the bullet.

[Anonymous No.14398615]

>>14398452
Honestly, I’ve already configured a system which would effectively conceptualize and implement a real, working, AI which is in comparison to a functional human brain, but it’s not worth telling any of you guys, or anyone, really. I’ll just bide my time using the hidden cross-sections if knowledge I’ve been able to find in order to better my own lot first. Sucks to suck, peasant scum.

[Anonymous No.14398618]

>>14398595
Thats right. We're going to sit on this rock until the sun eats it.

[Anonymous No.14398630]

>>14398615
>AI which is in comparison to a functional human brain
No, you didn't. You are a lying piece of shit, and you are most certainly going to hell.
>>14398618
We're going to die off by 2100 due to ecological catastrophes, and honestly, this isn't a big deal. Mankind is a heavily flawed species.

[Anonymous No.14398634]

>>14398595
>argh i'll kill you for speaking the truth
haha not if the augmented cyborgs get you first

[Anonymous No.14398647]

>>14398630
We’re already in hell, you big dummy. This is why I’m so much further ahead than anyone else currently alive. You guys are so far behind and so caught up in the spectacle that I’m not sure how to even explain it without creating massive amount of cognitive dissonance within your mind.

[Anonymous No.14398650]

>>14398634
You don't live in reality, and you will rightfully die in utter destitution, loneliness, and filth. This is because you don't truly possess a soul or any degree of real self-awareness or qualia.

[Anonymous No.14398684]

>>14398650
your violent outbursts helping you cope there little buddy?

[Anonymous No.14398698]

>>14398684
Yeah, we'll see who's right when WW III and other insane conflicts happen in 2039-2050. There won't be any transhumanist, singularity, or sci-fi-like "progress". This hype, from charlatans like Musk, is doing more damage than good. It is no different than waiting for Godot.

[Anonymous No.14398711]

>>14398595
incredibly based. faggots completely colonized by synthwave mixes and sci-fi aesthetics. it's unbelievable the damage media has wrought on our grasp of reality

[Anonymous No.14398714]

>>14398452
>“The Hedonistic Imperative outlines how genetic engineering and nanotechnology will abolish suffering in all sentient life. This project is ambitious but technically feasible. It is also instrumentally rational and ethically mandatory. The metabolic pathways of pain and malaise evolved only because they once served the fitness of our genes. They will be replaced by a different sort of neural architecture. States of sublime well-being are destined to become the genetically pre-programmed norm of mental health. The world's last aversive experience will be a precisely dateable event.”
>"To understand the world - both its formal/mathematical and its subjective properties - sentient organic life must bootstrap its way to supersentient Full-Spectrum Superintelligence. We need to find ways to navigate all possible state-spaces of qualia, including all possible first-person perspectives, and map them - initially via the neural correlates of consciousness - onto the formalism of mathematical physics."
>David Pearce, author of The Hedonistic Imperative
thoughts?
apparently this stuff is all the rage with all the computational 'x' people at Stanford and the like.

[Anonymous No.14398734]

>>14398711
This is your opinion, and your opinion has been warped to believe this to be the correct interpretation of the reality at hand, and your opinion is effectively wrong.

[Anonymous No.14398744]

>>14398698
you admit the only thing stopping it is collapse of industrial society or an extinction event. there will indeed be a war between transhumans and the original. guess who wins? your shrieking about muh qualia muh stemlords into the void can't stop or slow the inevitable.

[Anonymous No.14398751]

>>14398714
>technically feasible.
It is not feasible.
Genes exist in regulatory networks and altering a few expressions and so forth could have disastrous effects.
Nanotechnology has many hard-limits. Don't even get me started with how little progress is made in soft robots. I read 3-5 pages filled with differential equations of just one finger tapping; it dealt with dimensionality.
I wish every single scientific progressivist would die. I am being unironic here. Anyone who feels any enchantment with all of these scientific lies should get the bullet.
Anyone who worships GEIST above LEBEN. He or she does not deserve to live.
Go ahead and talk about science in a pragmatic and REALISTIC sense, but spare the bullshit, you fucking retarded naive and progressivist trash. You are just as stupid as cannibals of Western Africa in your own inane ways.

[Anonymous No.14398760]

>>14398734
Doubt it. Turn off the synthwave mix and go outside.

[Anonymous No.14398764]

>>14398744
>you admit the only thing stopping it is collapse of industrial society or an extinction event.
No, improve your reading comprehension faggot. There are serious "hard limits" and technical difficulties in bringing forth your transhumanist fantasies. Do not listen to idiots like Musk, Kurzweil, or Markram. Listen to people Francois Chollet or Peter O' Keefe instead. I was actually able to predict Keefe would win Nobel Prize fyi.

[Anonymous No.14398767]

>>14398760
But I live out in the woods? My population density is ~5 people per square mile.

[Anonymous No.14398777]

>>14398767
And you still got suckered in by the commoditization of cyberpunk.

[Anonymous No.14398786]

>>14398764
>Peter O' Keefe
Excuse me, I meant to say John O'Keefe. I was able to predict he would win Nobel Prize two years before he got it. His research in theta phase precession of rat hippocampus was interesting but not sci-fi level shit.
Anyways, I didn't enjoy that stuff because sacrificing animals and decapitating plus collecting their brains emotionally traumatized me.

[Anonymous No.14398797]

>>14398786
>theta phase precession of rat hippocampus was interesting but not sci-fi level shit.
hippocampal theta phase precession, grid and place cells, and etc involved in encoding spatial memory*. Here is your precision, dumb piece of shit. He was the one who really got into nitty-gritty details of that in one published journal article.
Only the mesoscopic scale of brain activity level matters, but we barely understand it except on a very rudimentary level with rats. We know barely anything about rhythmic brain activity in the human neocortex on a mesoscopic level

[Anonymous No.14398809]

>>14398751
>everyone not dogmatically committed to my steady state worldview must die
You're not making the humanoid case very well, you homicidal psycho. I know you're the neuroscientist from the other thread who wants theocracy. You got issues

[Anonymous No.14398820]

>>14398777
And what does that entail?

[Anonymous No.14398826]

>>14398809
I am also a computer scientist.
All transhumanist, naive sci-fi scenarios, and etc. are bullshit, and I know based on nuanced scientific knowledge. They are bullshit. If you want to justify scientific "progress", then you must do so in a more realistic and less quixotic manner. Do you not see with your hype you are proving anarcho-primatives and others right? You are not going to advanced brain-computer interfaces, space colonization, autonomous AGI (basically sentient AI), and etc. Moreover, most drugs for neurological reasons cause more damage than good.

[Anonymous No.14398831]

>>14398820
Everything the pro-transhumanists itt are arguing. Sci-fi fantasia. Complete delusions.

[Anonymous No.14398880]

>>14398809
>>14398826
Also, don't get me started on the damage many pesticides and preservatives cause. BPA, for example, causes negative epigenetic changes to babies in the wombs of pregnant of women. Also, look up Atrazine controversy. These are just few examples out of many. Many of these pesticides have research funded by corporations pushing for certain results.
Stuff like Big Pharma is real too.

[Anonymous No.14398905]

>>14398826
>stop it you're only proving the anprims right
oh no! :^)
your denial of scientific progress altogether though is some bizarre romantic escapism and too silly to merit a response. it's just like your dreams of a theocratic state that will re-enchant the world and respect your raw feels unlike these tyrannically materialist stemfags. it's based in your alienation rather than, as you keep insisting, a cool headed and unflinching thorough examination of the evidence.

[Anonymous No.14398947]

>>14398905
>you're a product of your conditions, but I'm not

shut the fuck up faggot.

[Anonymous No.14398949]

>>14398905
You are interpreting scientific "progress" in a kind of quasi-Hegelian way, which simply does not work. I even told you there will be progress in some areas like self-driving cars, better face recognition, better game AI (e.g., AlphaGo), and much more. However, there won't be any sci-fi bullshit like autonomous artificial general intelligence (basically sentient AI), advanced brain-computer interfaces, space colonization, or any of that sci-fi bullshit. I said this multiple times. How old are you, and do you have any scientific background of your own outside of pop trash?

[Anonymous No.14399065]

>>14398831
This is your opinion.

[Anonymous No.14399072]

>>14399065
His opinion is correct. Some opinions are good, others are trash. Your opinions are most likely trash.

[Anonymous No.14399134]

>>14398949
that science is a double edged sword doesn't refute anything but the most naive worship of progress. you've thus far projected all kinds of shit on me, from hegelism to denial of qualia. keep jousting with the strawmen inside your own head though if it makes you feel better.
>>14398947
>noooo that's an ad hominem you can't just accurately diagnose the pathology behind certain posts

[Anonymous No.14399158]

>>14399134
>>noooo that's an ad hominem you can't just accurately diagnose the pathology behind certain posts

feminine tactic.

[Anonymous No.14399177]

>>14399134
>the most naive worship of progress
That's precisely what you're doing.
>from hegelism to denial of qualia
You are completely missing my point. Since you love empirical science so much, take some reputable Coursera classes online, email professors or PhD students, or whatever. As I've told you, I am involved in two STEM fields, and most of the hype regarding them is delusional. That's not to say they are completely worth. I am more ambivalent in that regard.

[Anonymous No.14399186]

>>14399177
>completely worth
completely worthless*

[Anonymous No.14399195]

>>14399158
as if arguments were free floating in a vacuum and not all too human artefacts

[Anonymous No.14399197]

>>14399072
His opinions are wrong and emotionally charged.

[Anonymous No.14399205]

>>14399197
People who believe in sci-fi nonsense typically have no real scientific background of their own.

[Anonymous No.14399206]

>>14399197
>nooo you can't feel things and not be a sterile gormling like me nooo

[Anonymous No.14399210]

>>14399205
Just because you’ll be dead when most of this manifests doesn’t mean it won’t happen.

[Anonymous No.14399280]

>>14399210
All I know is it won't happen during our lifetimes. There is no way to verify what will happen after our lifetimes though. I lean towards no, but how can we determine who is correct after we're dead?
BTW, almost everything you hear about quantum computers is an exaggeration. As my physics major friend says, "There are some pretty severe problems with quantum computers - probably the biggest one is thermal noise which limits the fidelity of your output singal. I've seen the results of some cutting edge research with quantum computers, and they are not very impressive so far."
Most sci-fi stuff surrounding Neuroscience and Computer Science are exaggerations. That much I can tell you.

[Anonymous No.14399342]

>>14399280
Well, that’s because our current model of the universe, quantum physics, is wrong. There is also a fundamental flaw in the way we conceptualize the brain and the way it operates; this should be completely obvious to anyone who’s studied behaviorism and neuroscience. When you account of these two errors, you’ll see and understand why building a fully sentient AI and many of the other things we’d wish to have seen invented by now haven’t been.

[Anonymous No.14399432]

>>14399342
We do know some important things about the brain like how thalamocortical oscillations are necessary for consciousness. We don't know the precise mechanisms sufficient for consciousness because neuroimaging techniques are poor at revealing mesoscopic dynamics, which I did delve into. In fact, I explained how I was able to predict one researcher who would win Nobel in biology.
>our current model of the universe, quantum physics, is wrong.
Of course, but good luck figuring out what is right. It is much easier to critique current models than it is to find out what is correct or most closely conforms to the processes we see in nature. This is why I believe all empirical science exists for instrumental purposes rather than uncovering the nature of reality.
>why building a fully sentient AI and many of the other things
I recommend reading this. The brain has little in common with computers:
https://qr.ae/TWTSo2
Paul King's responses on Quora tend to be good. I recommend looking more over his stuff if this topic interests you.

[Anonymous No.14399489]

>>14399432
they are recently coming up with 3D neuromorphic architectures using memristors. that's already a step above traditional silicon which is 2D.

[Anonymous No.14399560]

>>14399432
>thalamocortical oscillations
I will admit, I am not familiar with that term. However, that being said, I am familiar with what gives rise to self-consciousness in the sense of metaphysical concepts, which I believe most in modern science tend to lack, as they are so caught up in the empiricism and spectacle of it all. The critical question always remains unanswered: how do you program desire into an object? I believe that when this answer can be supplied, we will have crossed the threshold into developing not just sentient AI, but literally anything, in comparison to God itself.
>Of course, but good luck figuring out what is right
I will not claim I know the exact definitions of what is right, but there are people who have “enlightened” my understanding. My critical observation has always been what is the defining characteristic which links all material things in existence? Again, don’t get so caught up in the empiricism of it all and define it down to more arbitrary particles like quarks, neutrinos, and such. If you continue down that line of reasoning, you’ll just continue to miss the point. Follow the old adage: as above, so below.
>Paul King's responses
Truthfully, I’m not interested. A TL;DR would be appreciated.
>The brain has little in common with computers
This is a bold statement that requires a large amount of corroboration as to what exactly the brain is, and what a computer is. You’ll find the answer may surprise you.

I could elaborate on further upon my own knowledge, but that’d spoil your treasure hunt and truthfully, it’d hinder my own pursuits of selfish fame.

[Anonymous No.14399589]

>>14399560
Based schizo posted

[Anonymous No.14399635]

Bump, any good books about philosophy of science.

[Anonymous No.14399647]

>>14399589
Flattering, but I just like to lift heavy things and chase the Logos.

[Anonymous No.14399933]

>>14398647
elabórate faggor

[Anonymous No.14400056]

>>14399342
>Well, that’s because our current model of the universe, quantum physics, is wrong
No it isn't

[Anonymous No.14400178]

>>14400056
>dark matter
>dark energy
>gravitons
>There is no complete quantum field theory of gravitons due to an outstanding mathematical problem with renormalization in general relativity
Just lmao. Way to buy into the bullshit.

[Anonymous No.14400214]

>>14400178
Incomplete does not mean wrong.

[Anonymous No.14400392]

>>14399635
btw you're inside a simulation, sv

[Anonymous No.14400417]

>>14400214
The language of quantum gravity is a false and perverted interpretation of the Logos. It’s wrong and blasphemous to continue parroting this failed language. Muh “we just don’t have the right math” screams to me that if the most intelligent of mathematicians can’t solve the most crucial aspect of unifying the field, than its flat out dead in the water.

[Anonymous No.14400448]

>>14399560
>This is a bold statement that requires a large amount of corroboration as to what exactly the brain is, and what a computer is.
>Truthfully, I’m not interested. A TL;DR would be appreciated.
I'll just copy-paste his response:

"The brain is neither analog nor digital, but works using a signal processing paradigm that has some properties in common with both.

Unlike a digital computer, the brain does not use binary logic or binary addressable memory, and it does not perform binary arithmetic. Information in the brain is represented in terms of statistical approximations and estimations rather than exact values. The brain is also non-deterministic and cannot replay instruction sequences with error-free precision. So in all these ways, the brain is definitely not "digital".

At the same time, the signals sent around the brain are "either-or" states that are similar to binary. A neuron fires or it does not. These all-or-nothing pulses are the basic language of the brain. So in this sense, the brain is computing using something like binary signals. Instead of 1s and 0s, or "on" and "off", the brain uses "spike" or "no spike" (referring to the firing of a neuron)...

Internal to the neuron, everything works via biochemical pathways, which are somewhat similar to analog. Neurons also perform internal electrical signal integration in an analog fashion. Analogously, the digital logic gates used by computers are implemented internally using transistors and resistors, which are also analog.

This recording of neural spikes over time shows that the spatiotemporal pulses of the neural code looks a lot like digital signaling:

...

The bottom line is that the brain processes information using a representation strategy that is neither analog nor digital. It is a different type of computation, involving circuits and networks composed of spiking neurons. One of the central tasks of neuroscience is to figure out how this information processing paradigm works."

[Anonymous No.14400450]

>>14400417
>using religious terminology in a scientific discussion
Go back to /x/ schizo. You gave yourself away

[Anonymous No.14400457]

>>14400448
There is absolutely no reason that neural architecture could not be replicated if it was sufficiently understood

[Anonymous No.14400460]

>>14398452
reminder that the chinese room argument is undefeatable

[Anonymous No.14400464]

>>14399489
Interesting. Isn't a lot of memristor research being done in China? I actually don't know about that stuff much... but it does seem to be better for mimicking the human brain.

[Anonymous No.14400469]

>>14400457
Not with computers binary logic. Artificial Neural Networks have little to do with the brain's neural networks. Maybe memristor could work though. Traditional computers can't, and deep learning models are limited in many ways.

[Anonymous No.14400490]

>>14400469
neural networks should probably be called something else in all honesty. there is a growing push to research alternative computer architecture for the reasons you brought up. not necessarily to try and replicate true cognition, but because many tasks would be more efficiently handled by a computer using 'less precise' logic like biological brains do. I don't want to downplay the importance of deep learning and the current AI wave, though. this type of software is a huge step forward in computing, but not because it anything to do with literal artificial intelligence.

[Anonymous No.14400542]

>>14400450
Like I have stated before in this thread: I am not well read on the empirical scientific data and terminology, which has thus far not served any purpose other than showing us that empirical science, as we employ it, is wholly flawed and incapable of representing that which is wishes to represent.
>>14400448
>The brain is also non-deterministic
Argument flawed right at this moment. This is why you retards need to study behaviorism and animal psychology, as well as a total understanding of Logos metaphysics.

[Anonymous No.14400562]

>>14400460
The spergs in this thread can easily tear to pieces and I'm sure they will.
Chinese room is a running gag on /sci/. Nothing more embarrassing than philosophers trying to play at science like they understand any of it.

[Anonymous No.14400579]

>>14400490
I take it you're not this idiot here?:
>>14400542

[Anonymous No.14400591]

>>14400579
Honestly, dismiss me all you want. You’re going to feel really stupid in the next decade when my theory full btfos everything you’ve ever known.

[Anonymous No.14400607]

>>14400579
I'm not the schizo. don't respond to crazy people who think their semi-theological-philosophy self-published science breakthroughs are legit

[Anonymous No.14400613]

>>14400562
>anime

Then refute it.

[Anonymous No.14401519]

>>14398452

Any book recommendations to understand OP's pic? I've read some Quine, Kuhn, and Tegmark.

[Anonymous No.14401522]

bumping a nice general

[Anonymous No.14402768]

>>14398714
Mankind does not strive for happiness; only the Englishman does.

[Anonymous No.14402953]

>>14400562
>moron actually thinks /sci/ is smart
Bait? /sci/ is way dumber than /lit/, I used to browse it almost exclusively as a stem undergrad

[Anonymous No.14403039]

>>14400542
>yeah I don’t know any science but it’s definitely all wrong haha
dumb schizoposter

[Anonymous No.14403053]

>>14400460
Meaning what?

[Anonymous No.14403342]

>>14400460
>Searle asserts that there is no essential difference between the roles of the computer and himself in the experiment. Each simply follows a program, step-by-step, producing a behavior which is then interpreted by the user as demonstrating intelligent conversation. However, Searle himself would not be able to understand the conversation. ("I don't speak a word of Chinese,"he points out.) Therefore, he argues, it follows that the computer would not be able to understand the conversation either.
Neither does a Chinese toddler speak Chinese. They learn it by emulation, trial and error. After several age-related brain upgrades and much punishment-vs-reward conditioning, the Chinese young lad speaks fluently. Exchange a hieroglyph for a tonical syllable, the basic spoken Chinese language unit.
There is nothing to 'understand' in the very first sounds your parents throw at you, you create meaning by cataloguing parents reactions to sound you hear or say via association. Then producing or hearing sounds (different parts of the brain are in charge of these) tags associated experience. The reactions catalogue is basically 'the processing manual' from the experiment.

Honestly, it sounds like the stereotypical bullshit that passes for philosophy, using vague images or words instead of precise predictable language. Because then suddenly you have the immediate 'does not follow' reaction instead of imaginary mind-wandering.

[Anonymous No.14403353]

>>14403342
Except a baby eventually grows into an adult that doesn't mechanically repeat strings of sentences, what's your point?

[Anonymous No.14403368]

>>14403353
The argument seems to imply that if you do not emulate a splitting image of interpretation, then the exchange is lost and unable to be translated back in forth in any meaningful way. Of course, this is bull shit and completely reductionist in scope of what a computer is and how exactly does one “computer” communicate with a “non-computer”.

[Anonymous No.14403503]

>>14401519
Bas van Fraassen 'Arguments Concerning Scientific Realism' and Stathis Psillos 'is structural realism the best of both worlds'

[Anonymous No.14403942]

>>14403503

Thank you

[Anonymous No.14403987]

>>14403342
this is a reformulation of what searle calls "the systems reply". simply put you cannot move from syntax to semantics. the child assigns semantics to formal language, which AI would never be able to do

[Anonymous No.14403996]

>>14403987
AI in its current conception never could, and Searle seems to imply that will always be the case, which is highly reductionist in scope.

[Anonymous No.14404027]

>>14403996
searle thinks that AI cannot merely by virtue of it being AI. he thinks that machines and computers and so on could eventually think. but they would need something other than just code in order to accomplish that

[Anonymous No.14404083]

>>14398595
>>14398630
>>14398650
>>14398698
>>14398711
>>14398764
>>14398786
Stop reading Zizek, you insufferable queer.

[Anonymous No.14404155]

>>14404027
No shit. This is why the standard model of AI and the way we conceptualize autonomous computing is inherently flawed.