Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
Science

Mathematical Problems For The New Age 353

Thanks to David A. Madore who wrote to us regarding seven math problems that have rewards of one million dollars. The project is being done by the Clay Mathematics Institute, and is modeled after Hilbert's list for the 20th century which was announced also in Paris, but in 1900. Read more for more information from David.

"The Clay Mathematics Institute has just (May 24-25, 2000) organized a big meeting in the Collège de France in Paris, to celebrate the hundredth anniversary of the International Congress of Mathematicians meeting in August 1900 (also in Paris) during which the great David Hilbert (a serious candidate for the greatest mathematician of all times, perhaps after Gauss and Euler) announced his list of 23 celebrated problems for the XXth century.

First the Clay Mathematics Awards were given, one two Laurent Lafforgue for his proof of the local Langlands correspondence, and one to Alain Connes for his work on von Neumann factors and noncommutative geometry. Then a list of seven problems for the third millennium was revealed by John Tate and Sir Michael Atiyah. If you solve any of the following problems, you win $1,000,000.

The problems are:

  • The Riemann hypothesis: prove (or disprove!) that all the non-trivial zeroes of the Riemann zeta function lie on the critical axis (real part of s = one half).
  • The conjecture of Birch and Swinnerton-Dyer: prove (or disprove!) that the algebraic rank of an elliptic curve over Q (the rank of the group of its rational points) equals its analytic rank (the order of cancellation at 1 of its L function).
  • Is P=NP? In other words, is it possible for a deterministic Turing machine to solve in polynomial time problems which are solved by a nondeterministic Turing machine in polynomial time, or, on the contrary, is the traveling salesman problem truly "hard" in the sense that no polynomial-time algorithm exists to solve it?
  • The Poincaré Conjecture: prove (or disprove!) that any simply connected topological manifold of dimension 3 is homeomorphic to the three-sphere.
  • The Hodge Conjecture: prove (or disprove!) that on projective algebraic varieties, all Hodge cycles are rational combinations of algebraic cycles.
  • Yang-Mills theory: develop a mathematical foundation for the quantum theory of Yang-Mills fields; specifically, account for the "mass gap" hypothesis.
  • Navier-Stokes equations: prove (or disprove!) that the Navier-Stokes equations have smooth solutions for all positive times, given reasonable initial conditions.

Naturally, if you solve any of these, you get to be famous, too.

During the meeting we even got to hear a recording of Hilbert's voice speaking his famous "Wir müssen wissen; wir werden wissen" ("We must know; we shall know") just a hundred years before. It's also engraved on his tomb in Göttingen. Cool. "

This discussion has been archived. No new comments can be posted.

Mathematical Problems For The New Age

Comments Filter:
  • by Anonymous Coward
    Nice one, I could never remember my 9 times table. Now all I have to do is think 9*x is the same as xxxxxxxxx / 12345679

    Genius!

  • by Anonymous Coward
    Actually the job market for phd mathematicians has improved in the last couple of years. From about 1989 to 1998 it was very bad. In a typical year about 1200 phds in math are awarded in the US; a large fraction (2/3rds?) are able to get academic posts, although many of these are temporary. For a reasonably talented mathematician (i.e., can complete a phd at a good school), an academic career is a reasonable expectation. You can then obsessively spend your time working on strange problems (although the advice to stay away from famous problems is wise, and you will have to earn your keep by teaching).

    Speaking of strange problems: My favorite is this one. Pick a number $x>1$ not an integer. Look at its powers $x^n$, $n>1$. Look at the fractional part of its powers. Any pattern? It is known that for "almost all" numbers (in the sense of Lebesgue), the fractional parts are uniformly distributed. But no-one knows for specific numbers like $e$ or $\pi$. No one knows what the distribution is for $x=3/2$. But check out $x=\frac{1+\sqrt{5}}2$, the golden ratio. The fractional parts all are either 0.000... or 0.9999 (alternating) after awhile. Algebraic numbers with this property are called Pisot numbers. No one knows if there are transcendental numbers that do this.

  • by Anonymous Coward
    Here is another famous mathematical formula which combines the contants e, i, pi, 1, and 0. e*i*pi*1*0=0
  • by Anonymous Coward
    Reminds me of an extra credit problem I heard about on a test.

    This is extra credit. Answer all parts. No partial credit will be given.

    a) Prove or disprove Goldbach's Conjecture.

    b) Prove or disprove the RSA Assumption
    (i.e. breaking RSA is hard)

    c) Prove or disprove : P = NP.

    The way I heard it, some people took it seriously
    and complained to the prof that it was "too hard..."
  • by Anonymous Coward
    Take a typical NP search space, and start numbering the states (breadth first search works well, as it will number infinite search spaces) Now the proof that finding the minimal (or maximal, ordinal) state given N of them takes O(N) time if the states contain no information about each other(they are uncorrelated). So the problem reduces to finding one such search space so that there are 2^N states, and you can prove that the states are statistically uncorrelated. That problem cannot be solved in polynomial time. Finding such a search space is an exercise left to the reader.
  • Most likely you have a much better chance at winning the lottery, unless you don't compete in the lottery in which case both chances equal nil :)

    Accually your odds of winning the lottery (By which we mean the biggest jackpot in the largest drawing, since smaller lotteries are winnable) is almost exactly the same wither you buy a ticker or not. Remember you can find the winning ticket on the sidewalk. For normal statistics you can ignore those events as insignificant, but winning the lottery is so statisticly unlikely at you have to factor finding the winning ticket it, which is only slightly smaller. (Order of mangnitude yes, but not significant)

    The odds that I will come up with any solution in my lifetime are very low as I currently stand - I don't think about math in general much. If I decided to solve one problem my odds of doing so in my lifetime increase dramaticly. Of course deciding that I want to solve one of those problems means that spend all my freetime on it. I'd sell my boat since I wouldn't use it, move into a tiny house (preferabbly near a major university so I can access their library and professors... perhaps even become a professor though as others have said I'd be unlikely to make tenure) I'm smart enough to make reasonable progress in a solution if I dedicated my life to it. I may or may not find a solution, but I have a chance.

    Dedicate your life to being the big winnder in the lottery doesn't depend much on your brains. While it is true that I can increase my odds of winning the lottery, to do so significantly would mean leaving cheaply (again, no boat...) and making enough money that in the end I'd have more money saving it all then I'd get when I got the winning ticket.

  • Since I haven't seen any links posted to Hilbert's original list yet: a copy of his original paper [clarku.edu] is avaiable for those interested.
  • It is not the case that M (in the previous) post must be prime. It is possible for it to be divisible by some prime greater than pn.

    The smallest such example is 2*3*5*7*11*13 + 1 which is 59*509 (both of which are primes).

    Secondly, checking divisibility by primes is enough, because if some non prime N was a factor of M, then any prime that divided into N would also divide into M. So the only factors of N (and it must have non-unit factors, since it is not a prime) can be composites (non primes). Then apply the same process to them (i.e. they can only have composite factors) and so on. The problem is that your factors are always getting smaller and are always positive. Eventually you run out of positive integers, so the original assumption that N has only composite factors must be wrong.

    (OK, that last paragraph was a little more complicated that I was intending it to be.)

  • Einstein couldn't do simple math. He had to get someone to help him with geometry for his own theories...

    Nope. Replace 'simple' by 'esoteric non-trivial' and you come closer to real thing.

    Einstein was pondering about how to extend his theory of special relativity, which is suited for uniform motion to be extended to the case where acceleration happens.

    He was a genius in the art of thought experiments, and realized that if you are closed into an elevator without the means to look outside, you can't distinguish if you experience acceleration to the bottom of the elevator by a source of gravity or by someone accelerating/pulling the elvevator.

    More abstract, he concluded an eminent physical priciple:

    Dynamical and gravitational mass are the same!

    That the 'm' for mass that shows up in Newton's equation of motion is the same as the 'm' that shows up in the law of gravitation, the gravitational charge of materia so to say.

    He went one and found that this narrows the set of all possible physical laws down considerably. (Physical laws must be covariant).

    So he talked to his friend Marcel Grossman, that he should go to the library and look for a mathematical framwork that had this and that features.

    Like other people order pizza, Einstein orderd non-Eucledean geometry this way.

    Grossman could report to him that Bernhard Riemann, based on work of Carl Friedrich Gauss, had already developed a suitable theory a hundered years before. Differential Geometry on Manifolds in a modern speaking, Tensor calculus in more old fashioned formulation.

    That stuff is not easy. It is not standard curriculum in Physics and except for mathematicians, only some mechanical engineers in fluid and solid dynamics use that framework.

  • One of the truly remarkable things about Einsteins theory of general relativity is that is not ivory tower stuff, pleasing some physicists, but actually this is put to use in ordinary every day devices:

    The GPS system (basically a bunch of high precision atomic clocks that send down their time) uses general relativistic corrections to achieve its high position resolution!

    On the other hand I know no application of the theory of strong interaction (Quantumn chromodynamics - think quarks) of any kind.

  • Correct. I was referring to the "naive algorithm", with is O(2^n), and therefore NP.

  • In one of my CS classes, we had to write an algorithm to solve the "subset sum" problem. The naive solution is to, basically, count from 1 to 2^n (where n is the number of elements in the list). Then, use i'th bit in the counter as boolean values to determine whether or not to include the i'th value in the list to add into a guess. I.e., for count == 1, see value[0] equals the target. For count == 2, see if value[1] equals the target. Count == 3 -> value[0] + value[1].

    The algorithm I came up with used recursive subsets of the (sorted) listed, as well as pre-calculated knowledge of the sum of the first i elements. Try as I might, I could *not* get an exponential-time result from sample input. The teacher "proved" that my solution was still NP, but it was kind of brushed over. I'm not saying that I wrote a P algorithm, but I certainly never saw proof to the contrary. Is there a (friendly) group of people who would be willing to look at such things from a lowly student? I'd be perfectly willing to be proved incorrect, but neither do I want to just give up simply because I'm not a world-famous scientist.

  • You're correct, of course. What I meant to say was that if an algorithm with a polynomial runtime exists for a problem, then the problem can't be NP.

    Can it?

    We didn't go nearly as in-depth as I would have liked, so please understand that I'm approaching the discussion with an "interested amateur" point of view.

    On a side note, my exposure to this kind of theory prompted me to join the ACM's [acm.org] SIGACT [acm.org]. That is, the Special Interest Group on Algorithms and Computation Theory. It's a great place to be if you like this stuff.

  • He said double the largest known prime and add two. That way the result is gauranteed not to be a prime (it's even!) and it is not the sum of two primes.

    This and the original post said the sum of two primes. Otherwise we could just keep adding one to our heart's content.

    What may also apply is that the primes may be positive and negative. I don't know.

    Now it's just a matter of finding two prime factors which sum to a number twice the largest known prime plus two :-)

  • He said double the largest known prime and add two. That way the result is garaunteed not to be a prime (it's even!) and it is not the sum of two primes.

    wrong, its not going to be the sum of two known primes.

    Oh that's just splitting hairs... I actually meant "the" two primes. You're absolutely correct, but I think it was quite clear that I didn't think it a solution for the problem when I said that all you had to do was calculate which two primes sum to that total.

    What may also apply is that the primes may be positive and negative. I don't know.

    All primes are positive by definition.

    When my statement is qualified with the words "I don't know", you certainly can't fault me for claiming to be an authority on the matter :-)

    But thanks for the correction.

  • take the number 12345679 (there's no 8)
    multiply by a number that the sum is 9 and see:
    12345679 x 09 = 111111111
    12345679 x 18 = 222222222
    12345679 x 27 = 333333333
    12345679 x 36 = 444444444
    12345679 x 45 = 555555555
    12345679 x 54 = 666666666
    12345679 x 63 = 777777777
    12345679 x 72 = 888888888
    12345679 x 81 = 999999999

    who discovered that? is there a name to this thing? do you know any others funny values that make the same thing? is there anything to proved in that?
    --
    BeDevId 15453 - Download BeOS R5 Lite [be.com] free!
  • Nonononon...the solution to fermat's theorem was most definately not obvious - I didn't mean that one, I just meant some other proofs that I've seen. You're right - the way the guy tied everything together to prove fermat's last theorem was the genius in that proof, but any proof so long as to require several PhD's years to read the hundreds of pages it contains is not what I would call obvious. :)

    Ah, and the old 0.999999999... thing. Most people don't realize that it is 1 until you tell them this:

    What does it mean for one number to be equal to another number? Well, if x = y, then x - y = 0. In this case, what's 1 - 0.9999999999999...? The answer is 0.000000000000....(infinite number of 0's). Since 0.000000000.... = 0, they are indeed the same number.

  • Alan Turing was a gay, British, atheist.

    David Hilbert was a heterosexual (even went after his grad student's wives) German. I don't know whether he was a Christian or not though.

    Cheers,
    Ben
  • Many number theoretic results, if they are undecidable, cannot be proved to be undecidable.

    Others can.

    The Goldbach problem is an example of something which if it is undecidable, cannot be proved to be undecidable.

    P=NP is something that could theoretically be undecidable and be proven undecidable.

    In both cases the potential counter-examples can be enumerated (ie placed into a list). The difference is as follows. There might be an algorithm that actually works for P=NP which you cannot actually prove works. However you can always prove one way or another whether a specific integer is the sum of two primes.

    Cheers,
    Ben
  • The maths and mathematical history in GEB is all correct, as far as I am aware.
  • by SimonK ( 7722 )
    There is no largest prime number. This is one of the very few things about prime numbers that can be proven.

    If there were a largest prime number, there would be finite list of primes you could write down. If you took all of these numbers and multiplied them together, then added one, you would get a new prime number.
  • Dear ___Anonymous Coward___,

    The first error in your proof occurs on page ____1____, in line _____3____.

    Sincerely,

    Joe

    That is, you make a fundamental error about the implications of reducing a problem to another problem. If I reduce a problem to a polynomial problem using a polynomial transformation, I know I have a polynomial algorithm, but you propose doing the reverse: "reducing" the problem to an exponential problem. That proves nothing; after all, I can provide you with an exponential sort algorithm (try every permutation until you find one that puts the values in order). The problem you transform to may not be the least expensive problem.

    Hence you do not prove that P != NP.

  • A quick Google search:

    http://www.mathematical.com/mathgold1.html

    Now I'm no mathematician, so I obviously didn't check to make sure they carried all of their ones, but at a glance, it looks ok. (How's that for an endorsement?)

    Cynic
    http://napalm.firest0rm.org/
  • sure, it's first order. but there are plenty of undecidable first-order statements, so this proves nothing about this problem being decidable or not.
  • Oops...it does make a difference to your system of mathematics, but either way, the system generated by either assumption cannot be shown to be inconsistent.

  • Actually these japanese guys were Taniyama und Shimura and the conjecture is called the - guess -
    Taniyama-Shimura conjecture.

    It is about elliptic curves and modular forms (or whatever that is in englisch).
  • Having worked on the Navier Stokes equation for about three years, I can say that this is a very interesting problem. Actually, the Navier Stokes equation is the equation governing an incompressible flow - that is, it is an equation that very closely approximates the flow of water. Thus to gain any understanding of this equation is to make a lot of progress in understanding turbulance. This will have great practical applications.

    I know a little about some of the other problems, the Poincare conjecture, P=NP, and the Riemann hypothesis. Although these problems are hard, not too many people work on them. Basically, for an untenured professor, these are career destroyers. If one works on these problems as obsessively as a solution will require, you can kiss goodbye to tenure (unless you actually solve it).

    In contrast, there are many many researchers the world over working on the Navier-Stokes problem, from graduate students to top mathematicians (like Charles Fefferman who wrote on the claymath website). One may not solve the real problem, but there is progress to be made, and many other easier problems to work on. One can make a reasonable career, while still having a dream of fame.

    ----

    Another poster said that whoever solves this problem will probably already be very rich, working for some rich institute. This really is not the case - probably the solution will come from a moderately paid university professor. The money will be very handy, although then his/her career will be so certain that he/she will be comfortable without it. In any case, the motivation will probably be prestige rather than the money, but the amount of money raises the prestige.

    Another poster commented that there are no Good Will's out there. I personally did not like the movie. I think that there are unfound genius's out there - Ramanujan is an example. But noone can solve these kinds of problems without being to some extent obsessive. Good Will did math as an afterthought, and no person can do great work like that. All the great genius's of history, in all creative endeviours, were not just brilliant, but also worked extremely hard.
  • Wasn't this posted months ago? Or was that a different problem? I seem to recall another math problem that a million dollar reward was offered for recently...

    --
    grappler
  • Just found the article (funny, I looked for it BEFORE I posted and didn't see it).

    It's the Goldbach Conjecture [faber.co.uk], from two months ago [slashdot.org].

    --
    grappler
  • by Frijoles ( 16015 )
    Time to go buy that 'Dummies guide to math' book.
  • So if we send x out to infinity, y will still be bigger by exactly 1.
    You seem to be laboring under the rather common misconception that infinity is a number (at least, not in the ordinare "real" number system). It is not. You cannot set x equal to infinity, so you cannot have y=infinity + 1.

    If you're still having trouble, try this thought exercise: Let's construct our infinitely repeating number like this: write 0.01, then insert 0's just to the right of the decimal forever.
    "Zero followed by a decimal point followed by an infinite number of zeroes followed by a one" does not represent a real number.

    If you're interested in this sort of thing, I REALLY think you ought to study calculus (and I don't just mean take the class). The sorts of mathematical and logic errors that you are making peaked in the mathematical world in the 19th century and were put to rest when calculus was finally put on a mathematically rigorous foundation. I think if you understood the foundations of calculus you'd have a better idea of why you are wrong. If you don't understand those foundations, there is no way I can really describe to you why you are wrong. It'd be like trying to explain pointers to a non-programmer who's doesn't speak your language.
  • I think anything else is hand-waving, which can often get you into mathematical trouble, like when you try to reorder with alternating series.
    I'm not a mathematician either - never finishe dup the graduate degrees. But it looks to me like his proof could easily be made rigorous in the form which he describe. Define the notation so that 0.999... and 1.000... represent the appropriate infinite series, define the procedure for subtracting and borrowing, and presto! The outline he's given becomes fairly rigorous.
  • Poincaré argued, however, that he used a cyclical arguament to show the principle of induction.
    Which is why the principle of induction is generally introduced as a postulate in most courses that start with Peano's postulates.
  • So Q is a number theoretical conjecture of the form "All A have property B." A is a member of a countably infinite set - positive integers greater than three, for example. This is an important point, as it means that a suitable programmed computer would find a counterexample, if it existed, in a finite number of steps.

    1. If Q is provably true, then there exists a series of statements that constitutes a proof of Q.

    2. If Q is false, then there exists a counterexample.
    2a (contrapositive of 2) If a counter example does NOT exist, then Q is NOT false. (ie no counterexample == true)

    If Q undecidable, then there does not exist a series of statements that constitute a proof, and there does not exist a counterexample. But wait! If there's no counterexample, the conjecture must be true. Proving the conjecture's undecidability thus produces a contradiction - you have proved it true AND you have proved it undecidable. So the conjecture is not undecideable.

    Either there's a flaw in my logic, or yours. Which, and where? Or were you merely being more general than I?
  • I beg to differ. After that infinite number of 0's is the digit "1" at position infinity+1 (or -infinity-1 depending on how you index) - that is, perhaps a more helpful notation would be 0.00000...001, which != 0.
    Ummm - no, that'd be the case only if 0.999... = 0.9999...990. The sequence of nines does not terminate. The sequence of zeroes does not terminate either. Granted, the poster you are replying to is giving an intuitive demonstration rather than a rigorous proof, but you are proposing mathematical nonsense.

    More true would be to say that if x=1 and as y approaches 1, x-y approaches 0 so we can select a y to get x-y arbitrarily close to 0.
    This is true for a finite number of nines following the decimal point. The logic does not hold up if you make the leap to an infinite number of nines. In other words, 0.9999.... != 0.999....9990.

    If I remember my infinite series correctly (it's been 10 years since calc 2 for me), 0.999 represents

    oo -----> (think 8 on its side)
    ---
    \ 9
    \ -------
    / 10^(-n)
    /
    ---
    n=1

    I've long since forgotten how to sum the series rigorously, but I haven't forgotten the result; 0.99... = 1.

    In summary, the arithmetic of infinite series is intuitively nonobvious, and slightly different from the arithmetic of
  • This whole "contest" is fascinating. On the one hand, fewer and fewer people are able to make a living as mathematicians. On the other hand, to solve these kinds of problems, one has to throw oneself at them whole-heartedly. It does seem like the money should come up front to support those capable of such solutions; somewhat of a mathematical patronage. But, the question then becomes one of purpose. I love the notion of solving these kinds of problems, but to whom is it worth a million dollars? As the problems become harder and more abstract, solvable by fewer people, the tangible benefit seems to tend toward naught. It is a little akin to mountain climbing expeditions. It serves a corporation no direct benefit to sponsor these trips. Are we going to make the transition from contest to corporate-sponsored mathematicians?
  • In your algorithm you say to "count from 1 to 2^n" where n is the size of the list. That is the problem right there, looping 2^n times is exponential (2^n is an exponential, not a polynomial). Polynomial time algorithms are allowed to loop n times, or n^2 times, even n^100 times. But you're not allowed to loop 2^n times in a polynomial time algorithm.
  • Time to go buy that 'Dummies guide to math' book.

    Ooooooh, now you've done it... IDG are gonna have to close down slashdot now... I hope you're happy!

    <G>

  • "Isn't there a "largest" known prime number?

    "Known", yes. Just plain largest, no. Proof:

    Let n1, n2, n3...nx be all known prime numbers. Construct M = n1 * n2 * n3 * ... * nx + 1. Is M a compound number (i.e. non-prime?)? Well, it isn't factorable by any known primes (it would always leave a remainder of 1). So we have two cases:

    Case 1: M is prime. But it must be a previously unknown prime and it is clearly larger than any previously unknown prime. Therefore there exists another largest prime.

    Case 2: M is not prime, but isn't factorable by any known prime. Therefore it is factorable by some unknown prime larger than all known primes.

    Since this process can be repeated indefinitely, there must be an infinitude of primes.
    --
    Have Exchange users? Want to run Linux? Can't afford OpenMail?
  • Hey, just got back. Why are you guys all so old?
  • When, again, does the 20th century start?

    :)
  • Yes, thank you very much, my school [www.ens.fr] (translate: college) happens to be where Bourbaki's office is (and, essentially, his "birth place"), and I've read a significant portion of his "Éléments de Mathématiques".

    I fail to see how that is relevant. Bourbaki was (were?) not interested in logic. His (their?)treatment of logic and set theory in his first book ("Théorie des Ensembles" - whose publication was long delayed as it remained in the state of a "fascicule de résultats") is sketchy at best, unelegant, and, in essence, as short as he could arrange it, just enough to provide the foundations for the other books.

    Anyway, that is old. Currently, there are only about two groups doing logic in France: around Krivine in Paris and around Girard in Marseille (I think). Girard is the inventor of linear logic, so that's a clame to fame, but he's more the exception than the rule: logic is held in disrepute, if not contempt, in France (and I much regret it).

  • Professor Andrew Wiles was present at the meeting (he made a little speech to introduce Laurent Lafforgue and later John Tate - although the latter really needs no introduction). I can assure you he did not turn red.

  • Most likely you have a much better chance at winning the lottery, unless you don't compete in the lottery in which case both chances equal nil :)

    These problems are _hard_.
    IMHO the most notorious one is the Riemann Hypothesis. Some of the best mathematicians have been working for years on that one. It has very broad implications (but so have most of the others).

    For example, one can show that there's a strip around this critical axis in which all non-trivial zeroes are located. The smaller this strip, the more accurate are our estimates on things like 'the number of primes not larger than x'.

    If the Riemann hypothesis were disproved, a lot of mathematics would immediately become invalid (many papers assume the validity of the hypothesis to prove deep results). In this respect it is quite similar to the Taniyama conjecture (from which the truth of Fermat's last theorem followed, together with lots of other theorem based on this conjecture).

    Should the two number-theoretic conjectures (Riemann and Birch Swinnerton-Dyer) be proven in the first half of this century, I suggest that number theorists concentrate on the abc conjecture in the 2nd half of this century.
  • The single fact that Fermat's last theorem is of mostly historical value. However, it was proved by showing that a large parts of the Taniyama conjecture hold (in fact, by now the full Taniyama conjecture has been proven). The Taniyama conjecture is important in the theory of elliptic curves, and a lot of mathematics has been built on top of it (even though for a long time it was not clear that the conjecture was actually true). The theory of elliptic curves nowadays has applications in areas from cryptography (factorization of integers) to theoretical physics.

    The 7 problems at hand have similar implications. For example if we know that the Riemann Hypothesis is true, we suddenly have very precise estimates on for example the growth of the number of primes. Also, the Riemann Hypothesis has generalizations in other areas of mathematics (some of these generalizations have been proven by now).
  • But the text enter box at /.is too small to post it.

    George
  • Grr I forgot about > being munched. That last line should have read:
    "y prime" is "for all j, j > y or exists k such that jk < y and j(k+1) > y".
  • How many are left?
    ... the only one I know about being solved is FLT.
    ---
  • Hilbert a gay British atheist? I think you have him confused with someone else...
  • Goldbach's conjecture is first-order. There is no quantification over proerties or sets of numbers in it.
  • divec's original post is simply ignorant, as I pointed out in this [slashdot.org]
    post. Goldbach's conjecture is first-order, but it is a first
    order sentence of a theory which cannot be given a primitive recursive
    axiomatisation. A lot of people here are confused about language
    vs. theory and decidability vs. completeness: back to the textbooks, I
    think.
  • 1. Elevendy-four :QED
    2. 42
    3. 69 (Ooh, sexy.)
    4. Both 7 and 11 (it's a quantum problem)
    5. 14 cats + 1 hedgehog * Larry Storch
    6. No
    7. E=mc^3

    Please have my check ready, I will be over to get it after "Full House".
  • Send me a binding offer, signed and all. My email is posted on ./; send me yours; I'll send you a fax #; you fax me your offer. I'm dead serious. Not that I'm hoping to ever collect that $1M from you. I just want to frame the offer and hang it from my wall.

    You'll get the proof anyway.
    --

  • Isn't there a "largest" known prime number? Take double that + 2 and isn't it disproven?
    Of course I'm the furthest thing from a mathematician and I was really happy when I knew the answer to "the longest side of a right triangle" on Who Wants to Be a Millionaire, so I'm probably wrong but that's what comes to mind if I recall some of the math I had way back when.
  • Yes. Godel showed that the continuum hypothesis (CH, which says aleph_1 = |R| = pow(aleph_0)) is consistent with ZFC -- therefore, it cannot be DISproved. Cohen in 1965 or so showed that CH is not decidable within ZFC: that it cannot be proved, either.

    [ZFC is set of axioms used as the basis for a (the?) standard formulation of set theory: you start with these 9 axioms and rules of logic and all of your set theory follows.]

    Therefore CH is an axiom that is independent of ZFC -- you can choose to adjoin it to your axioms or not, and you get a different set theory depending on whether you use it or not. Some might say that you choose whether to 'believe' CH -- more accurately, you choose to use CH depending on what kind of model of the real world you want your theory to be.

    [btw, one can also prove that the 8 axioms of ZF are independent of C (the axiom of choice). So as above, you can choose whether to adjoin it to your set of axioms and make ZFC or just stick with ZF.]
  • What if you can prove that there's no proof for whether it is decideable or not?

    I claim to have a proof that this situation is impossible. There could be a flaw in my logic.

    suppose that the original statement is decidable. then there's a proof that it's true, or there's a counterexample. therefore its decidability is definately true (and therefore decidable).

    now suppose the decidability isn't decidable. then the original statement can't be decidable (see above). therefore, the original statement is not decidable.

    --

  • I thought this through a while back, and here's what I came up with. I'd appreciate if someone could tell me where it breaks down.

    We know that the reals can be mapped onto the reals from 0 to 1. So, consider the following sequence of numbers:

    000000000000000...
    100000000000000...
    010000000000000...
    110000000000000...
    001000000000000...
    ...

    Just a progression of binary integers. Okay. Now, consider these as binary expansions of a real number less than one:

    0.10000000000000...

    So we have a mapping from the above to all reals less than one. We can also consider this a mapping onto the power set of the integers:

    010010010000000... {1, 4, 7}

    So we have the same progression mapping onto both sets of interest. But, aleph0 of the reals (terminating fractions) will be encoded twice. It seems like we could in some way "subtract these out", since subtracting a lower order of infinity from a higher order should still give the higher order. Also, the diagonalization argument that Cantor used doesn't work, because the number generated is:

    0.1111111111111111111111... = 1

    But we said that we are limiting it to numbers less than one.

    But, each of the bit sequences we are using to encode a subset or a real also encodes an integer. So the above shows that:

    |reals| = |power set of integers| = |integers|

    Which we know is false because everyone says so. So, where is the flaw?

    --Kevin
  • What does it mean for one number to be equal to another number? Well, if x = y, then x - y = 0. In this case, what's 1 - 0.9999999999999...? The answer is 0.000000000000....(infinite number of 0's). Since 0.000000000.... = 0, they are indeed the same number.

    Actually, there's a much easier to understand proof that 0.999... is equal to 1. What's 1/9? 0.111111.... What's 2/9? 0.22222.... So what's 9/9? 0.999999..... Of course, 9/9 is 1.

  • Actually, R is the set of all Reals. C is the set of all Complex numbers (which include the Reals, Imaginary numbers, and sums of the two).
  • Although it's interesting to see an academic body offer a lot of cash to solve these problems, they're asking the wrong crowd. The people who have a shot at solving these problems are probably the same people who are offering the cash. If they can't do it, how could the non-mathematicians do it? There are no Will Hunting's out here.
  • Math is always news for nerds, because programming is very much like the activity proving theorems - you go for simplicity, generality, and ABOVE ALL, conceptual elegance.

    And, actually, the Is P=NP question is vitally important to any sort of reasonable algorithms work. Travelling salesman is one problem that's easy to explain, and even easier to tie up huge amounts of resources with. But it's equivalent to natural language processing, for instance, and a lot of useful AI applications. Reducing any of those problems to polynomial time would be the same as reducing all of them, and the fact of the matter is that that is almost certainly worth far more than a measely $1e6. More like $1e(8 or 9).

    So, yeah, these are incredibly hard problems. And like all of the real stumpers in math, they begin with one question: where to start?

    Ushers will eat latecomers.

  • Sorry - I mean to say "It is widely believed, but not proven, that the problem of factoring a large number is NP-hard". That is, it believed to be in NP, but outside of P.
  • Not true. It is widely believed, but not proven, that the problem of factoring a large number is NP. This difficulty (even if it is not NP) is what RSA depends upon. If P = NP, then obviously, there is a polynomial time algorithm to decrypt RSA encrypted messages. But if P != NP, we still have to prove that decrypting RSA is really NP.
  • Seeing how that nobody answered your question, let me try.

    Let n be the largest known prime number. Now is n+2 itself prime? (You have to prove/disprove this one)If it is, (just like 5 and 7) then you've just increased the largest known prime number by 2. Then it is just another instance of Goldbach's conjuecture.

    If it is not prime, things become more interesting. Is there some primes less then n, such that when you add them together, you get n+2? If you can prove that there isn't, then you have disproven Goldbach's conjecture. But chances are, there is, since we can do (for example) 3 + 3 + 3 + 3 + ... and use that hit a lot of numbers. Good luck finding that number missed by ALL such sequences ....

  • I just want to be able to prove (or disprove) that the amount I think is in my chequing account has a direct relationship to the amount that is actually in the account.

    I'd pay good money for that!

  • This is a trivial, solved problem. The method of generating functions solves this and a much more general set of enumeration and probability problems. Informally, a generating function is a function whose formal power series (to be distinguished from the more familiar Taylor power series) has coeffecients equal to the corresponding value of the series that the function generates. Wilf's _Generationfunctionology_ (yes, the title is real) explains this very well, much better than I could do, since I only have a cursory understanding of these methods. By the way, even the naive approach of enumerating all the possibilities is easily computing, in a certain abstract sense. The function is primitive recursive and can be computed in polynomial time using the general methods in Knuth's _TAOCP_. Volume 4, when it comes out, will go into more detail about these techniques.
  • Flow of incompressible fluids?!?! That's what the Navier-Stokes differential equations are for. Most mathematicians believe that the GRH is almost certainly true. There will be no new applied results that suddenly arise when GRH is finally proven. We already have a very good sense of the distribution of primes, but it is not possible to predict the next prime in any real sense (that is, a priori based on being given an arbitrary prime), except for actually finding it. The GRH offers insight into how the primes are distributed, but not where a specific prime would be; RSA depends not on the difficulty of finding large primes (this is already easy: Miller-Rabin SPPTs are entirely adaquate, use eliptic curve techniques if you wish to prove primality), but rather that of decomposing large composites into primes. GRH offers no insight into how to do this; even if it did, we can merely assume that it is true, without any proof, and use the corresponding results. Proving GRH!=breaking RSA.
  • I don't know about other problems BUT if anyone of you, guys know how to solve this one, I will personally get 10,000,000 into your pocket. If this problem is solved I will buy the solution and the rights to the solution for 10 million dollars US from ANYBODY as long as it is shown completely and with code example that this is doable.

    Why such extreme numbers? Well, basically, if the solution to this problem is found, then I can easily reduce the solution into ATM or into Turint Halting problem and by doing that I can basically solve every single protection algorithm on this planet in polynomial time (in no time - rather.)

    So, if any of you out there, who wants to make 10million USD, you should solve this problem and contact me.


  • Given a phrase of boolean variables, determine if any combination of values for variables makes the phrase true (Satisfiability)

    Given a set of numbers, and a target sum, determine if any combination of the numbers add up to the target sum (Subset-sum problem)

    Please excuse my ignorance on this, but this both seems easy if you can use really fsck bloated and slow moving code. I assume this won't be consider polynomial time? Could anyone explain what exact is 'polynomial time' is? How it is defined?

  • The flaw's in your logic.

    You're conflating what's usually referred to as "Proof theory" with what's usually referred to as "Model theory". By Godel's Completeness Theorem (not Incompleteness, Completeness -- they're different), if a statement is provable in first order logic from a theory, then it is true in all models of that theory -- and vice versa. (If the statement is provable, then it's true everywhere.) 2a doesn't actually follow from 2. 2a follows (after a highly non-trivial proof!) from the stronger statement "If Q is false, then there is a counterexample in all models."

    Think about the axioms for the Theory of Groups. There are perfectly good groups in which multiplication isn't commutative. That means that the Abelian axiom (for all x and y, xy == yx) isn't provable from the standard axioms of group theory. It's not a proof of a contradiction, it's just a proof that a sentence isn't decided.

    A theory in which all sentences are decided is called "complete". Very few theories are complete, and, in fact, what Godel's incompleteness theorem shows is that any theory which codes up the notion of proof in a non-trivial way relative to itself -- pretty much, any interesting theory -- is, ipso facto, not complete.
  • There was a /. story a month or so back about the Goldbach conjecture being worth a million.

    At least I could understand that problem. These look hopeless. What is a topological manifold anyway? Whenever I hear that, I always picture mathematicians going down to the junkyard and picking apart exhaust systems... Q: Why is that math professor making all that noise and stinking up the place? A: Sounds like he has a hole in his topological manifold. :)

  • Modular forms are strange four-dimensional things.

    The Taniyama-Shimura conjecture was that every elliptic curve was a modular form. This implied that Fermat's Last Theorem was true in a roundabout way. Wiles proved Taniyama-Shimura and thus Fermat.
  • All even numbers greater than 2 are the sum of two prime numbers.

    Well, there is the obvious number theory rule that odd+odd=even. Since all primes >2 are odd...

    Try to read before replying....

    Jeroen

  • Emperically, with a computer.

    uh, no. Fermat's Last Theorem was solved analytically by Andrew Wiles, by hand, over a seven year period.

    The problem cannot be solved empirically, and neither can any of the others above (except maybe by counter-example) as they involve absolute proof of general cases

    And it was Fermat who wrote something like "I have a marvellous proof of this, which this margin is too small to contain"
  • I find it reassuring that in a time so obsessed with giving money away for just about any non-intellectual endevor, that a group has taken it upon themselves to reward those who accomplish with their minds, a trait which seems to be rather overlooked and undervalued these days.
  • I remember way back in freshman year when I took CS121 (Intro to Formal Systems and Computation)...

    The night before the final I posted a message on comp.theory asking for a proof that P=NP because it would greatly help on the problem set I was working on (namely invalidating the proofs I had to do to show that several different problems were NP-Complete).

    Well, Unfortunately everyone on the newsgroup took me seriously and there was some angry responses along the lines of: "Are you stupid? Your going ot fail your final tommorrow if you don't already know that this is the biggest unsolved porblem in computer science..."

    The best was that a few weeks later (after the grades were turned in, thank god), my TA noticed the post and E-mailed me in disbelieve asking if I had really posted that question...
    I don't kow how she knew it was me since I only signed with my initials...
  • Come on people!!!! The minute we start questioning whether we really need to know something is the minute we cast off our Linux boxes and head back for the caves!!! Talk about asinine comments. I thought we figured out a long time ago that scientific (including mathematical) research is justified in and of itself. Sheeesh!!!
  • >> Math is always news for nerds, because programming is very much like the activity proving theorems

    Actually proving theorem and programming _are_ the same things under certain constraints (basically that you don't prove that the [the property being false] is false in order to prove that the property is true). This is called the Curry-Howard isomorphism and is the general idea behind many formalizations of programming.
  • by Black Parrot ( 19622 ) on Thursday May 25, 2000 @05:17AM (#1049093)
    I don't see anything about the famous and all-important karma maximization problem.

    --
  • by FascDot Killed My Pr ( 24021 ) on Thursday May 25, 2000 @05:21AM (#1049094)
    More than once, in fact.

    Method 1: Follow story link. Cut random text and past to /..
    Method 2: Create post such that title = "Why this is good for Linux" and body contains random text.
    Method 3: Create post such that body contains "I am post anonymously because (insert random reason here)."
    Method 4: Post something critical of Jon Katz (only works temporarily, karma rapidly falls off after 5 minutes).
    --
    Have Exchange users? Want to run Linux? Can't afford OpenMail?
  • by David A. Madore ( 30444 ) on Thursday May 25, 2000 @03:36PM (#1049095) Homepage

    No, it's not available on-line. (Even after hearing it, I would really like to get it a copy.) Maybe by asking Professor Alain Connes (he seems to be the one who found the tape) very politely...

  • by anatoli ( 74215 ) on Thursday May 25, 2000 @06:37AM (#1049096) Homepage
    If I prove to you that Halting Problem is not in NP, will I get my $10M?
    --
  • by bifurcator ( 80430 ) on Thursday May 25, 2000 @06:57AM (#1049097)
    But the exact "value" of c, in relation to other transfinite cardinals was not determined. In particular, its relationship to Aleph1 (which I believe is the cardinality of the power set of Aleph0) is formally undecidable.

    You're pretty close there. But not quite. The cardinality of the power set of Aleph0 is C (the cardinality of the set of all real numbers). Put simply, Aleph 1 is the cardinal that comes right after Aleph0. C looked like a pretty good candidate for some time. C is a cardinal, and it comes after Aleph0. The question was, does it come right after Aleph0, or is there another cardinal in between Aleph0 and C? If my memory serves me right, an American named Cohen proved that you cannot answer 'yes' or 'no' to this question within an axiomatic system like ZFC (Zermelo-Frenkel plus the axiom of choice.)

  • by Anonymous._.Coward ( 119202 ) on Thursday May 25, 2000 @05:25AM (#1049098) Homepage
    Excellent, now I don't have to spend any time making up the finals exam questions for my math students.

    Prof. AC

  • by jabkie ( 121992 ) on Thursday May 25, 2000 @05:30AM (#1049099)
    don't forget that proving Goldbach's conjecture is also worth a million bucks [faber.co.uk]

    the /. story is here [slashdot.org].

    the conjecture itself (for the link-following-impaired) is:

    Every even number greater than two is the sum of two primes.

    Enjoy!
    --

  • by dillon_rinker ( 17944 ) on Thursday May 25, 2000 @09:09AM (#1049100) Homepage
    Dear ___Tebriel___,

    The first error in your proof occurs on page ____1____, in line _____2____.

    Sincerely,

    Dillon

    When the math department where I went to school received a crank proof, this form letter would be stapled to the front, a grad student would fill it out, and the proof would be returned to its sender.+
  • by David A. Madore ( 30444 ) on Thursday May 25, 2000 @02:53PM (#1049101) Homepage

    All right. I posted the story, now here are a few comments of my own.

    The problems are all fairly old. The Riemann hypothesis is the oldest, as it dates from around 1860 and it was already part of Hilbert's 1900 list. Surprisingly, almost no progress has been made toward its resolution (except for Deligne's proof in the generalized local case, aka the Weil conjectures, but that isn't really the same problem). The Poincaré conjecture is also rather old. The study of the Navier-Stokes equation is probably as old as the equations themselves, but the mathematical foundations for a rigorous study are more recent.

    P=NP is the newest problem, coming as it does from computer science, and also the most likely to be of interest to Slashdot geeks. John Tate rather messed up (IMHO) in his presentation of it (he obviously didn't think it was very interesting), and I gather many mathematicians believe it isn't very rigorously refined (certainly, many French mathematicians, with their usual contempt for logic and computers science). In fact, it is perfectly rigorous and mathematicial. It is probably the easiest problem on the list, as the theory behind it is still not fully developped. For a start, I'd recommend Papadimitriou's book "Computational Complexity" for the present state of knowledge: it is very thorough, well-written and interesting.

    Personally, I'm convinced P=NP is undecidable (in a nutshell, my point is that for Turing machines with oracles, by judiciously choosing the oracle we can arrange so that P=NP or P!=NP, see Papadimitriou's above-mentioned book for a proof, and I don't despair of the existence of a notion of forcing on oracles of some kind). It would be a very weird situation (for Pi^1 propositions, as a previous poster pointed out, if it is undecidable then it is true - proviso the consistency of arithmetic which is the hidden assumption in the meta-proof that "unrefutable=>true"; but for P=NP it just looks... weird). I don't think any of the other statements are undecidable (if any of them is, we are light-years away from being able to prove it). Naturally, it is possible for the undecidability of a statement to be undecidable, and so on up to (ordinal) heights of staggering difficulty (though by a theorem of Tarksi, at some point the proposition must be decidable - little comfort in practice).

    Some have asked why Goldbach's conjecture (for which someone else is offering $1M) is not in the list. The fact is that it's not interesting. Every one of these seven problems, if solved, would have far-fetching consequences, and its proof would open vast new areas of research. Not so with Goldbach's conjecture. It is amusing because it is very simple, simple enough to be explained to laymen. But apart from a few additive number theorists, nobody is interested in it. Proving it would require ingenuity, certainly, and much technique in sieving methods and circle integrals; but nothing like opening new areas of mathematics. Don't waste your time on Goldbach's conjecture: it is highly technical and intellectually almost worthless.

    Anyway, getting back to the seven problems, some people I know have been rather angry that such old problems have been selected, not more recent stuff. In a way the list is rather "conservative". For example, the global case of the Langlands programme, which is probably the most ambitious programme in mathematics (at any rate, in number theory and algebra), the global analogue of what Lafforgue proved in the local ("functions field") case, was not selected. Perhaps also because it is thought to be too difficult...

  • by Chalst ( 57653 ) on Thursday May 25, 2000 @06:39AM (#1049102) Homepage Journal
    This is not true. Gödel's completeness theorem only applies to
    primitively recursive axiomatisations in first-order logic.
    Arithmetic, as Gödel's incompletensss theorem's show, cannot be
    captured by by such an axiomatisation. Goldbach's conjecture is a
    Pi^0_1 statement of arithmetic (a measure of the intrinsic logical
    complexity of a statement that includes the Gödel sentence), so as far
    as we know, we might not know of a formalisation capable of proving it
    to be true.
  • by sh_mmer ( 63202 ) on Thursday May 25, 2000 @06:49AM (#1049103) Homepage
    Goldbach's conjecture is a first-order statement...

    come on: Goldbach obviously is *not* first order. let me slightly rephrase Goldbach, and you will see:

    "for every integer n > 2, there exist positive primes p and q such that p+q = 2*n."

    that's the proper formulation, and hence the conjecture is apparently *not* first order!

    d'accord?

    sh_
  • by Tebriel ( 192168 ) on Thursday May 25, 2000 @05:26AM (#1049104)
    NP = P
    NP * 0 = P * 0
    0 = 0
    q.e.d.
    Gimmie my million!
  • by Uruk ( 4907 ) on Thursday May 25, 2000 @05:20AM (#1049105)
    Taking number theory in college and persuing math in my spare time, I've found out that sometimes the ones that look the easiest are actually the most nasty, bastardly, evil problems you've ever seen. Some may look easy, but you have to figure there's a reason why there's a $1 million prize on each question.

    Some of the toughest ones are the "prove or disprove" type problems - for example, before fermat's last theorem was proven, (some people are sticklers for long periods of peer review and say that it hasn't been officially proven yet) it was pretty easy with the use of computers to prove that fermat's last theorem held for all integers up through some ungodly number. But proving through "some ungodly number" != proving through all numbers. Induction is the name of the game.

    Math is always news for nerds, because programming is very much like the activity proving theorems - you go for simplicity, generality, and ABOVE ALL, conceptual elegance. The really cool part is that when you find a truly elegant, awesome proof, I almost always have that feeling of "damn! It was so OBVIOUS!" -- even though it wasn't, the elegance of the proof makes the concepts surrounding the program just gel perfectly.

    I hope to see solutions like that for some of these problems. I'm not really all that great at number theory, but I really enjoy doing it and watching the big boys do it.

  • by SgtPepper ( 5548 ) on Thursday May 25, 2000 @06:46AM (#1049106)
    Naw....

    "Unsolvable Obscure Math Problems For Dummies"

    THERE! NOW IDG will close down slashdot...

    shit...maybe i shoulda posted that AC...
  • Several of Hilbert's problems in 1900 were of the form "find an algorithm for such-and-such", where such-and-such was something like "determining whether a polynomial of arbitrary degree and arbitrary number of variables has integer roots." These problems were closed, but not strictly solved, since it was proved that there were no such algorithms to do what he wanted.

    I would find it extremely interesting if, in the 21st century, it were shown that many of these conjectures were also proved to be undecidable. It would certainly explain why some of them have taken so long to figure out.

    If you show that the conjecture is undecidable, you haven't proven or disproven it - so do you still get the million $$? :)

  • by smileyy ( 11535 ) <smileyy@gmail.com> on Thursday May 25, 2000 @05:32AM (#1049108)

    You're pretty close there. Aleph0 is the cardinality of the set of all integers, c is the cardinality of the set of all reals. Cantor's diagonalization proof demonstrates that c > Aleph0.

    But the exact "value" of c, in relation to other transfinite cardinals was not determined. In particular, its relationship to Aleph1 (which I believe is the cardinality of the power set of Aleph0) is formally undecidable. That is to say, it doesn't make any difference to your system of mathematics if you choose c

    It's been a while since I've studied this stuff, but I've found this a decent introduction to the subject [mathacademy.com].

  • by El Cabri ( 13930 ) on Thursday May 25, 2000 @06:04AM (#1049109) Journal
    Never heard of the New Economy ?

    Giving $1M when the problem _is_ solved is an old-fashioned, actual-results-based, old-economy way of doing things.

    CNNfn taught me that the New Economy way of doing it is :

    -- give me $10M now. I promise to hire the brightest people on earth.

    -- If I don't get anything done, I'll fill an IPO and screw some shareholders.

  • by YoJ ( 20860 ) on Thursday May 25, 2000 @05:37AM (#1049110) Journal
    Given the background of most readers of Slashdot, if you are thinking of trying to solve one of these problems the best one might be the P=NP problem. Most of the others require some high-level knowledge of mathematics or physics. The high powered theoretic scientists have all taken a look at P=NP. I think real progress will come from a non-specialist who has really original ideas about new algorithms.

    What you have to do: find an algorithm to solve any of the following problems in polynomial time.

    • Given a graph, find a cycle (loop) that visits every vertex exactly once (Hamiltonian cycle problem)
    • Given a weighted graph, find the least-weight path that visits every vertex (Travelling salesman)
    • Given a phrase of boolean variables, determine if any combination of values for variables makes the phrase true (Satisfiability)
    • Given a set of numbers, and a target sum, determine if any combination of the numbers add up to the target sum (Subset-sum problem)
    Most CS people think that P is not equal to NP. They might be right, but I think we have vastly underestimated the power of polynomial time algorithms.

    Problems that are in P can be solved for sure in polynomial time. Problems in NP have solutions that can be verified in polynomial time. For example, in the subset-sum problem it is easy to verify that the sum of a certain combination of numbers adds up to the target. It is harder to come up with the right combination. But is it so much harder that it cannot be done in polynomial time? Maybe not.

    -Nathan Whitehead

  • by Azog ( 20907 ) on Thursday May 25, 2000 @07:36AM (#1049111) Homepage
    P is the set of problems that can be solved in polynomial time; i.e. the set of problems for which polynomial-time algorithms are known.

    So what's a polynomial time algorithm?

    For some deterministic algorithm 'A', (like quicksort) which takes an input of size 'n' (the objects to be sorted), there is some formula that says how many computational steps must be carried out to complete the algorithm.

    A simple example. Suppose you write an algorithm to find the maximum number in an array by scanning through the whole array once. This will take time proportional to the size of the array. In mathematical jargon, this is O(n) "big-Oh of n" where n is the size of the array.

    You can have algorithms that are O(n^2) "n squared", O(n^5 + 3n^2 + 50n), or whatever. If the function of n is a polynomial in n, then the algorithm runs in polynomial time.

    But if your function is like O(2^n) "two to the n" that is not a polynomial in n, and you don't have a polynomial time algorithm.

    As a simple example, consider code breaking. In this case, n is usually the number of bits in the key. If you have a 128-bit key, there are 2^128 possible keys. If your "algorithm" is the obvious one of trying every single key one after another until one works, you will have to check O(2^n) keys. This is exponential time, and it's a very slow way to break codes.

    Anyway... A definition of NP is a little harder to understand. NP is not "non-polynomial" although people tend to think of it that way. What it really means is "non-deterministic polynomial". Non-determinstic algorithms cannot be directly implemented in computers. However, they can be converted into deterministic algorithms which run in exponential time.

    So: the P=NP question is: For all the problems where we know non-deterministic polynomial time algorithms, are there actually determinstic polynomial time algorithms that we just haven't been smart enough to figure out yet?

    Probably not. Why? Well, there's this "NP-completeness" thing. If you find some interesting problem that you can't get a (determinstic) polynomial time algorithm for, you can find a non-determistic algorithm, and then then prove your problem is "equivalent" to a known NP-complete problem. That means that if someone was able to find a deterministic polynomail time to solve YOUR problem, that same solution could be used to solve ALL OTHER NP-complete problems in polynomial time. And since hundreds of people haven't been able to solve them in years of trying, there probably isn't much point in trying to solve yours.

    At this point you talk to your thesis advisor and start working on fixed-parameter tractability, or polynomial time approximations, or other more esoteric computer science approaches to the problem...

    Now you know more than you wanted to.

    Torrey Hoffman (Azog)
  • by divec ( 48748 ) on Thursday May 25, 2000 @06:07AM (#1049112) Homepage
    Goldbach's conjecture is a first-order statement, so by Gödel's Completeness Theorem (which is *not* the same as his Incompleteness Theorem) there is either a proof or a disproof.

    A "first-order statement" is one which can be formulated without saying "For all X" or "Exists X" for any infinite set X. (It's ok to say "for all/exists x" where x is just a number, or a set element. Also I'm not stating this mathematically precisely).

    Goldbach is stated as "for all x there exists y,z such that y prime and z prime and y+z=x". "y prime" is a first order statement: "for all j, j > y or exists k such that jk y".

    So there *is* a (dis)?proof of Goldbach, just waiting for someone to find it.

  • by Kinthelt ( 96845 ) on Thursday May 25, 2000 @05:17AM (#1049113) Homepage
    We haven't even finished solving the all of the original 23 problems!

    One great example is Goldbach's Conjecture: All even numbers greater than 2 are the sum of two prime numbers.

    Mathematicians have tried for years to prove or disprove this one (or to prove it is unprovable), but still haven't come up with an answer. I think the thing that irks most mathematicians is the fact that it is so short and elementary.

E = MC ** 2 +- 3db

Working...