125 IQ Sufficient for Machine-Learning?

IQ is meme.
Just work it.

an IQ of 85+ is fine for anything man, dont use IQ as an excuse for being a loser and not following through or putting in work

/thread

Most of us have an IQ of 150+ as it's evidenced by the ubiquitous screenshots of Arch installations and Haskell fizzbuzzes.

OP, I'm not sure a machine has an IQ

Yes, sucessful and intelligent people tend to test well in IQ tests.

Does not change the fact that the quality (and consistency) of IQ tests is abysmal and the value calculated from them has 0 value outside of bragging material and comparisons WITHIN THE SAME TEST.

Noone on this entire planet will be able to tell you anything more precise than that you probably are above average intelligence from the shit your started this thread with.

*and the number calculated from them has 0 value

If you're asking whether your IQ is high enough for machine learning, you're probably not cut out for machine learning.

Not that machine learning is particularly difficult (unless you want to dive really deep), but because you need to be able to think critically about data and experiment with parameters and figure things out for yourself

Asking about your IQ suggests you don't have the mindset to teach yourself and feel comfortable learning and exploring things on your won.

This.

Your IQ test lied.

you sound like a total brainlet
you are the pompous fuck

I believe that there's a high correlation between IQ and financial prosperity up until about 130. Western society tends to fail those over 130. You can believe what you wish.

By your logic, a scientist shouldn't be a scientist if they choose to invest minimal energy reading papers or opinions about the topic before investing relatively huge amounts of resources in carrying out an experiment for themselves, which is a totally ass-backwards proposition.

I hate to break it to ya bud, but you're an idiot.

Says the genius who can't even type a proper sentence in what is, more than likely, his native language. Thanks for the chuckle.

Yeah, bullshit. You're telling me that someone with an 85 IQ is going to become a competent neurosurgeon? Give me a break. We're diving into "anything is possible" "let's place our bets on anomalies, since that's logical and fruitful" territory.

But it's true: IQ doesn't mean anything. I've known people with borderline-retarded IQ levels who were very good at certain mental tasks, and I've met people with IQs of 180 who couldn't do math or think logically to save their lives.

You learn shit through repetition if someone does something enough they're going to get better.

You don't even understand how IQ works, you will never accomplish anything in ML.

>anectodotal evidence means statistics doesn't work
>met multiple people with 180IQ(chance 1 to 20,696,863)
Brainlet.

I appreciate the civil conversation, but I'll have to disagree. Regarding the borderline-retarded IQ's, just because someone takes (a) test(s) doesn't necessarily mean they're interested, want to, or try. Anyone can score very low on an IQ test - just don't try. However, it's difficult for a low-IQ individual/group to score notably high, lest they are "lucky", which can be adjusted for by having them take multiple unrelated tests.
As for the "180 IQ" individuals, from the relatively small amount of research I've done on IQ tests, I've yet to come across one that claims it can test accurately above 175. Most seem to cap around 160 or 175. So, that leads me to believe that it's not unlikely that these people have just taken some laughable Facebook IQ test, the main purpose of which is to feed peoples' egos and gain site traffic.

I agree that most online "IQ tests" are obviously bullshit, for that very reason. But, I do believe that there are some that are relatively reputed, that can estimate you within your general ballpark range (+/- 7 pts).

>UI
ahahahahahahahahhahahahahahahahahhahahah

Ahaha

Oh, please, enlightened one, do show me the way.

I know, funny stuff. Bet you were making $50k at 19, weren't you?

Attached: 1522213615285.jpg (1000x664, 97K)

I’m gonna give you the "IQ doesn't mean anything and you can do anything if you believe in yourself" antithesis answer of "IQ doesn't mean anything and you can do anything if you believe in yourself" answers.
It isn’t going to be about your IQ. Not because “you need to believe in yourself”, but because of the companies that are going to be hosting this shit. It isn’t going to be the actual geniuses that are going to be working with Machine Learning, it’ll be people like pic related because of who will be leading in these fields. When all of this shit catches on to the point where getting a job in it is practical, I t’s going to be companies like Google, FaceBook, Apple, and Microsoft. If you’re lucky, you might have a company like IBM or a privately funded college project at the very best. All of those companies I just mentioned essentially hire based on diversity hiring and affirmative action. It isn’t going to be today’s brightest, it is going to be today’s most unique, which will stagnate progress unless you get the resources to do it on your own.

Attached: B3E83852-B9C0-4872-8A41-E29F81D1BFB1.jpg (1200x825, 148K)

ML is not hard. It's mostly importing libraries and trial and error to get your precision as high as possible on your training data. It doesn't even matter with the systems you build actually do their job properly.

125 IQ is enough to win the Nobel Prize.
Check Richard Feynman.

So we are better off making our own ML shit?

You chose the worse board to ask this question. The mere mention of IQ triggers all the redditors and underachievers on this board.

PhDs average an IQ of 120. At 125 you can expect to get a PhD if you put in the effort.

I would say go to college and major in applied math and computer science.

Machine Learning involves computer science, statistics, probability, a lot of linear algebra and a good bit of multivariable calculus.

This

Sure if you want to work as a data analyst but if you want to contribute to the field like OP it is a lot more advanced.

There is no such a people at China or Japan.

125 is sufficient for anything except perhaps making groundbreaking revolutionary discoveries. At this point hard work ethics is more important than IQ.

Groundbreaking revolutionary discoveries require large team of people a d huge resources nowadays. Intelligence is not really that important.

Top kek. Another """"high-iq"""" sperg thread.

IQ is a measure of pattern recognition. Machine learning is all about patterns.
125 is more than 80 which is about down's syndrome level but still average.

>and I've met people with IQs of 180 who couldn't do math or think logically to save their lives.
L O L

kek

Genuine ML PhD student. 140 IQ, was part of Mensa (no shit).
Study CS I'd you want to do ML, my mathematicians colleagues can't do shit

IQ doesn't mean anything for any field and is a broad measure that isn't even done right most of the time.

if you want to screen yourself I suggest finding a mentor over the summer or smth

you're either genius level or you're not. anything below 140 doesn't matter. get your head out of your ass and just give it a shot.

Feynman never got into Mensa and he was definitely more intelligent than 99% of people alive today so not sure what that says about the nature of IQ.

I suppose IQ is a valid measurement to distinguish between those who are genuinely slow and not, but I think using IQ scores as anything else is a serious flaw and will just limit you in thinking there are plateaus of human cognition that correlate with strict measurements of IQ.

>did a iq test online
>said i have iq of 70
>mfw

Interesting response.
"All of those companies I just mentioned essentially hire based on diversity hiring and affirmative action."
Are you sure about that? I'm sure that they make sure to fill their diversity quotas somehow, which will include a considerable amount of East Asian females and some Indian males, but I don't know about the SJW crowd having such a huge influence in these companies as you propose. My guess is that they're less of a minority than most other places, but still a minority.

My main concern was that I might be better off in just plain old programming than going over-my-head into a field populated in majority by 135+ IQ's.

Hm, perhaps that actually is the case. Right now I'm focusing on learning/becoming-very-fluent-in all the prerequisite maths, and haven't started in actual ML so as to avoid developing any bad habits (per advice online). Perhaps that may not be the best approach.

As to be expected. I've posted this on Quora, Reddit, HackerNews, and here (obviously most all leftwing heavy sites). These are all I know. Any recommendations? They'd be much appreciated.

Honestly, yeah. If you can.

>had a nepotist job GM at age 19
>can't figure out shit like this

you need to train your neural net a bit more,pal

You need at least 200 IQ to understand machine learning

go back to codecademy, retard

you're supposed to make the bloody machine learn, not yourself, you imbecile

Thanks for the encouragement. Right now I haven't dipped my toes into actual ML yet, but have been focused on becoming fluent in all the prereq maths.
In my experience, I've always excelled a lot more in self-studying than in a structured/classroom setting, although I haven't been in a physical school for about 6 years.
My main reasons for not going to school are to avoid unnecessary debt and save time and hassle.
Now, after studying maths for a little bit, and taking a few IQ tests that were a bit disappointed and disillusioning, I'm beginning to wonder if I should aim towards ML at all, or if my ROI would be much higher if I just went into programmer/"data-science", or that perhaps I should go to school after all, even though I'll be a bit late (I'm turning 21 this November).

I'm half Chinese, so although I can't read/write fluently, that route is somewhat open to me.

>I'm half Chinese
finally explained

>half
Please refrain from shooting up schools just because girls don't talk to you.

Oh come on, you mean to tell me that a "hard-working" 125 IQ is going to outdo an average-working or even slightly lazy 150 IQ in trying to make discoveries/inventions in the same field? I don't believe it.

Founding and running a business, yes (considered the 30 pt difference theory regarding sociability) but actual science/inventions? I don't think so.

True. Culturally a lot of inventions/discoveries tend to be pinpointed on one "genius" inventor, when in reality teams and decades of people were behind it (Tesla and modern electricity/A/C, for example), largely due to peoples' psychological need to idolize "heroes".

Yep, this is essentially the reason for my concern. Well put.

You're underestimating the amount of work needed to make discoveries, at least in CS / ML. You have to spend a lot of time writing non-challenging code, such as preprocessing data, plotting, and frameworks to test stuff.

I genius mind will be able to get better ideas and interpret results easily. But a lazy dude, even if genius, will struggle to cover the way between ideas and actual results.

Of course that if you stay in the more theoretical fields it can be easier. As finding proofs seems more about actual intelligcen and less about writing code.

I'm actually working on my PhD in physics. I said you need to be independent enough to just start teaching yourself shit, instead of asking if you're smart enough. Even worse, he asked if his IQ was high enough, even though IQ is a pretty bad metric.

Reading papers is doing research. It's teaching yourself. It's exactly what I was advocating.

I hate to break it to ya bud, but you're an idiot.

Hi Elliot, I thought you're dead already

Holy fuck, a legit supreme gentlemen. Can't make this shit up.

Attached: 1516328341634.jpg (1462x1462, 189K)

Thanks for the advice. This is the reason I started this thread, to see what people in/near the field had to say. Would you recommend I just avoid ML and go into programming? I'm not sure if I really want to be the only guy in the room with a sub-130 IQ, given that I even make it to the room, struggling to keep a job in the field.

And if so, would you recommend college as a necessity? At the moment, I'm studying my own ML curriculum (currently working on prereq maths).

If it came down to not even doing ML and just settling for regular programming, I probably would just self-teach, since I have two cousins who earn six-figures in their mid-thirties, who are completely self-taught.

Well, I definitely disagree with you about the "doesn't mean anything for any field" part, but I appreciate the suggestion regarding the mentor, and will keep it in mind.

I don't have an educated response to that. All I know is that Feynman is a pop-sci celebrity, which doesn't mean much (regarding actual scientific intellect).

It wasn't a nepotist job at all, you twat. I taught myself programming in 6 months online, spent 8 hours a day spamming resumes online for a week, got a few calls from Indian recruiters, got an interview, got an offer, worked for a year, and got a 25% raise before I quit. Boss literally told me to "come back" when I was done studying.

Meaning?

Good one.

It's machine learning, dumbo. You barely have to do anything.

Perhaps. Just depends on the intel/time-investment ratio then, I suppose.

>Meaning
meaning you're a diversity hire

Whatever you say, man. Sorry that you're so far up on your high-horse that you're unable to reach the keyboard to ask questions to people online in hopes that someone who may be more knowledgeable/experienced might provide some helpful advice.

>Professor
>Manhattan Project
>Nobel Prize winner

I think calling him a pop-sci celebrity is a bit of a stretch even for the usual retardation of this board.

Wow, creative.

Lmao.

be at the technical forum and talk about iq...

Sounds good to me.

The problem with IQ tests is whoever is giving you the test might not know enough about you to test you accurately.

I did shitty on IQ tests when I was teenage because nobody considered the possibility of me having dyslexia.

Only much later in life when I had a job writing down a bunch of truck and trailer numbers did I realise I had dyslexia.

I saw the number 6 flip upside down and become a 9.

Attached: 1530386583509.jpg (235x215, 7K)

>Sounds good to me.
Ok, so teach yourself ML in 6 months online and apply for a job. since you're a diversity hire you're going to the top of the list easily

Professor - Adequate room for nepotism
Nobel Prize - Adequate room for nepotism
Manhattan Project - like I said, I'm not even familiar with the guy and his work, other than that he's a famous Jewish Physicist with ~125 IQ.

>was part of Mensa

That's not something to be proud of, little girl.

Attached: spanish_pepe.jpg (300x180, 12K)

kek, and a poltard too. The lels can't stop coming.

It's an option. I'd prefer to ground myself with a solid skillset so that I'm not too dependent on a particular role in a particular company, though, which is why I created this thread in the first place.

If you are not familiar with his work it means you have never studied serious mathematics or physics so I'm not sure what it is you are telling other than trying to diminish his accomplishments because he was Jewish. He has contributed to the fundamentals of quantum physics something nobody who is there because of "nepotism" has. It's just asinine to claim.

High IQ is not relevant unless you're pursuing a PhD

What has iq to do with black mesa u spastic

Just fucking try it already.

Never said anything negative about him being Jewish other than that he is Jewish, but alright.

There's still heavy debate in the "serious" scientific community regarding the existence of dark matter, which to my knowledge, has never been firmly proven, so the act of printing a theory in a college textbook doesn't amount to much proof of "genius" in my book. Just my take, as a lay.

Anyways, I get the point about ~125 IQ's being able to make groundbreaking accomplishments/discoveries. Another example would be Watson with the helix. His IQ was in the 120's, I believe.

I'm not sure how much this relates to ML, though.

But IQ test are finding 9th/16th rectangle that fits the pattern. Wasn't dyslexia limited to writing and reading?

Why are you niggers replying to a legitimate sperg? It boggles my Nubian mind.