/sci/ - Science & Math

>>4725749 Also Albert Einstein failed at early school and had a hard time reading and repeating.

links

>http://en.wikipedia.org/wiki/Albert_Einstein#Early_life_and_education

link say

>Although it has been thought that Einstein had early speech difficulties, this is disputed by the Albert Einstein Archives, and he excelled at the first school that he attended.[12]

WTF

You will never fail school just because being "too smart". If you are smart then school is easy. That Einstein failed at school is a myth, something you would know if you actually, you know, read the wiki-article you linked to. He consistently got the highest grades possible in physics and math.

Fail because you are too smart? Nah

Fail because you are bored, lazy, depressive, preoccupied, etc etc? Sure

a person who hasn't realized yet that school is simply a training ground to install a ''work and repeat'' doctrine into students for future repetitive work in employment are really deluded.

There is an inclination of intelligence higher up in university level in some subjects... but up to mid university it is all ''book-smarts'' where those that are successful are those that rerad the same sentance over and over the most times.

you are trained to answer exam papers. You are not trained to be intelligent. Infact in most cases, you are trained to think inside the box, rather than outisde... to think certain predictable ways.

God yes it is possible to be too smart for school. I was wayyy to smart for graduate school, much less anything lower than that.

The problem is if you are too smart and have too many reasoning skills then most professors are just obviously stupid in comparison. Their assignments are stupid, their methods are stupid, and the topics they teach are stupid.

This is prevalent in my field (CS), but I know it is true to some degree in physics as well.

There is a list of instincts and behaviors that the most capable people derive regarding how one should behave towards others, learn, doubt etc. 99% of faculty fail to understand a large number of these.

For instance, I have no instinct to get mad when someone disagree's with me. My instictual response is something like "O rly? Do tell?" 99% of professors do not have this instinct. All but math/theory professors seem to be overly concerned with minutia when you can just generate general principals to solve the problems they are concerned with.

However if you are too smart for school then you should go off and do something by yourself. Like start inventing things and filing for patents.

>>4725773
Right..

>Like start inventing things and filing for patents.

tried that... came up with quite a few good ideas. Problem is, patents cost quite a bit of money and literally take years to follow through.

The only one you can be too smart for school is if you're God tier and some teacher is forcing you to 2+2 over and over.

Being underchallenged is a problem I think, and is one of the largest issues with the American system at the moment.

Christopher Langan is a complete idiot. Just watch him talk about his ideas. It's laughable horse shit. His theories are all fucking retarded and he says that people with high IQs are more capable of solving any problem than specialized and educated scientists because dey r smarter.

oh and another thing... If you have a truly "mathematical mind" (I tend to think of it more as a logical mind) then you see that a lot of commonly accepted proofs/results/beliefs in science have errors in reasoning in them. The halting undecidability proof is a good example. The halting problem can be solved for almost every kind of turing machine, except for the one constructed in the undecidability proof (and trivially so) that would also never exist as part of any other machine unless intentionally put there, and machines that solved questions we don't know the answer yet.

The latter class contains machines for which the answer is either yes or no, and is not a function of the machine itself.

Trying to explain stuff like this to a professor usually results in a response similar to that you would expect if you tried to debate someone's religion in their place of worship.

>>4725793
Just shut up you fucking aspie.

>>4725793 then you see that a lot of commonly accepted proofs/results/beliefs in science have errors in reasoning in them

give an example please

>>4725793
>Cannot into the simplest of logic.
Question:
>Can we (at least theoretically) determine for all algorithms whether they'll terminate.
Logic:
>Nope, here's a counterexample.
FinalState:
>Hurr durr that's a retarded counterexample.
...
>I have a mathematical mind
nope.avi.jpg.exe

>>4725799

I just did... the halting problem. Another is the idea that nothing travels ftl. By definition entanglement is something traveling ftl. There are infinite "Not even wrong" interpretations of the set of current observations (most scientists/faculty do not understand this, another of that list of instincts). We came up with one convoluted such interpretation of the bell experiments that results in things like logic breaking down... causality breaking down... etc...

However the most likely interpretation is that the underlying assumptions of general relativity fail regarding particles below a certain size that we have no or limited experience of.

This is basic epistemology- (limits of induction) you cannot make assumptions about something you have no experience of.

>>4725823
You are the worst kind of Aspie. Get over yourself; you're no where as smart as you think.

>>4725822

The problem is your inability to see the transparency in the halting problem machine's design thus allowing you to answer it's halting problem from the get go. It never halts because it just copies machines forever and ever. thus allowing you to answer by virtue of the fact that you know a priori having designed it a certain way.

Not everyone can think at this level, which is what makes so many people susceptible to poorly reasoned proofs like this.

>>4725838

I am not an aspie. I have no parietal occipital.

>>4725823
Have you even considered the possibility that you simply don't understand the halting problem and elementary physics?
I'm no expert on the latter, but you definitely displayed a profound lack of understanding of the former.

>>4725840

Lol I'm cringing in embarrassment for you over the internet.

I hope that fuels your ego and makes you feel even more superior and act this way even more so others can share in the joy that laughing at you brings.

>>4725845

Do you know how many drooling butthurt morons asked Einstein the same question?

Understanding this simple principal has nothing to do with understanding physics... that is the sad part about the whole thing. It is a universal principal of human understanding. You can never ever say something like "Nothing can go faster than light". Why? Because any rule you derive is based on assumptions, which are based on experiences, and your experiences do not extend to everything.

While physicists over the years have constantly gone back and said "Hmm it appears we were wrong", Epistemologists have been /facepalming over and over again thinking "Idiots..."

The truth is people like you are not cut out for the highest level of intellectual investigation. You are nothing more than a cog in a machine.

>>4725855
Reported.

Another good example is Cantor's diagonolization proof. It never ceases to amaze me how putting up a chart can divert the regular mathematician/scientist's attention away from what is actually going on.

Infinite sets do not have a size. Numbers are generated by our minds... perhaps to describe our surroundings, but still. There are not more reals than rationals... both are infinite and you will die before listing all of either.

The real set simply grows faster than the rational set, having basically any way you can think of to generate additional reals for the same set of naturals.

>>4725855

Posts like yours were one of the reasons I stopped coming to this board for over a year.

I don't know if you are trolling, it is sadly impossible to tell on here these days.
Regardless, your posts are obnoxious. Desist and/or leave.

>>4725855

If your level of thinking is so much higher than that of the world's leading computer scientists and your own professors then how have you applied it to advance your field?

inb4 >I don't care about that I don't wish to share my genius they are not worthy and I won't tell you either but trust me it's really cool I'm so smart

Langan does nothing but ramble about his shitty theories and his IQ. I doubt Leonard Susskind has Langan's IQ, but he sure as hell is a lot smarter than him.

Anyways, bottom line, school is nothing more than a bunch of people stuck in a certain way of thinking because they are dependent tribals who cannot think for themselves.

There is a huge difference between a person who can invent a new branch of scientific inquiry and someone who just scores A's at school. If you are in the first group, you hardly have the patience to suffer the school system.

In my area, the ideas prevalent are so backwards and wrong it was a complete waste of my time. In my area I was amazed at how far people will go in terms of memorizing convoluted bs approaches rather than stepping back and realizing there is a more optimal approach that is simpler than the status quo approach.

>>4725889

Then why are you not an accomplished scientist?

>>4725868
I have the general intelligence algorithm..

>>4725855
So now you're Einstein?
The halting problem is proven to be undecidable. Not understanding the proof does not equal being too smart for the proof.
You are mixing up different concepts. There are machines that do not halt, and for which it is decidable that they do not halt. There are machines that halt, and it is decidable that they halt.
There are also machines of which it is not known, and cannot be known, whether they halt.
It's a simple matter of enumeration of universal turing machines, and enumeration of inputs.
For each combination of machine and input, mark as 1 if it halts and as 0 if it doesn't. Let's refer to this as h(machine, input), which is 0 or 1.
(This should be possible if halting is decidable.)
Then, consider the algorithm e that computes g(i), for input i; g(i) will be 0, when h(i,i) = 0 and g(i) will be undefined, and e will loop forever when h(i,i) = 1.
Now, tell me, what is g(e)?
If g(e) = 0, then it must've been the case that h(e,e) = 0. However, program e halts on input e (remember that g(e) is the outcome of the program e, if it terminates). Hence h(e,e) must be 1. Contradiction.
If g(e) is undefined, then h(e,e) = 1. However, we have defined that e loops forever in that case. So h(e,e), which represents whether e terminates, should be 0. Again, contradiction.

Where's the error?

>>4725893

Then where can I read about it? Have you published it online or in an academic journal of any kind?

Will you tell me about it?

Or are you just some horse shit faggot who has no idea what he's talking about and tries to act superior online when in real life if you acted this way actual smart people would laugh at you and you would feel stupid and get butthurt?

If you can't produce ANYTHING to back up your claims, you are almost certainly a pathetic shithead who thinks he's a genius. It's OK, there are tons of people like you. You can find them at the bookstore in the science section trying to start conversations about their genius with anyone who is unfortunate enough to pass by.

I used to be a genius likr you, but then I took a

>>4725901

Not understanding why the proof is faulty does not equate to the proof being sound.

The error is that you (or rather Turing) trivially added the contradiction in the creation of h() and g(). This triviality is perceived by me, thus it can be perceived by the correct implementation (associative) of h().

One of 2 things are true about h(x,y):

1) It concerns itself with x's handling of y
2) It does not need y, because the answer does not depend on the input.

g(e) is an instance of 2. It is always the case that g(e) would loop forever if put into h(x, y) instead of h(x). An associative version of h can recognize e, and can also overcome rice's issues with being able to tractably identify g(e) hidden inside another program.

>>4725920
Tell you or I'm lying? Could you be more transparent. I am not about to give the GIA away for free to any idiot who wants to hear about it. I am getting ready to submit it for patent, and have concerns about doing even that. It can basically be used to genetically engineer viruses, or psychologically engineer serial killer computer programs. It is not likely to do something wrong by accident but it is likely to give a single individual a lot of power.

Even if I have faith that nothing bad would happen why would I give it away for free...

>>4725971
Associativity is h(h(x,y),z) = h(x,h(y,z)). How is that relevant.
Did you mean distributivity?
Have you considered that there operations that are not distributive, and that are computable? You can't simply assert that every algorithm is distributive.

>"But he grew up in poverty and says he was beaten by his stepfather from when he was almost six to when he was about fourteen.[6] By then Langan had begun weight training, and forcibly ended the abuse, throwing his stepfather out of the house and telling him never to return.[7]"

>"After earning a perfect score on the SAT[6](despite falling asleep) "

>"by his mid-40s had been a construction worker, cowboy, forest service firefighter, farmhand, and, for over twenty years, a bouncer on Long Island."

holy shit, Chris Langan rules. Confirmed alpha badass

>>4725979
Because us laughing at your brainfarts may be your only shot to get rid of your delusional sense of grandeur.
>Misunderstand proofs
>Proofs must be wrong
>2012

>>4725991

... no. I am talking about h() being an association rule data mining algorithm. As in, if you see a trivially non-halting program, then just put does not halt as the answer.

>>4725997
He proved that there is no implementation of h that works for ALL inputs (= all programs and all program inputs). You're saying:
>But it works for some inputs (namely association rule learn algo's, with some input programs).
Wooptidoo. You do understand the difference between universal and existential quantification don't you?