Missing Things

Showing posts with label mathematics. Show all posts
Showing posts with label mathematics. Show all posts

20 December 2012

Lockhart's Lament

I don't know if you've noticed that I'm really bad at this 'updating the blog on a regular schedule' thing? There's not even really much of an excuse for it; it's not as though I'm working on more important things (well, I am, but I'm spending a lot of poorly-prioritised time on less important things, too). I'll have to try to get back into the habit.

Anyway - I try not to talk about myself too much on this blog because it's really not about me so much as just me thinking about things. (In case that motif wasn't already obvious on the blog background!) But I'm making an exception, because there's something I absolutely HAVE to gloat about at the moment. So, personal anecdote first; if you get lost or bored with the mathematics, skip to the bold text below where things get less anecdotal again!

For a little background - my current college transcript contains references from four different schools already and I'm only halfway done. The problem this has led to is that I took basic classes like Calc I and II at one school, then transferred to another school that accepted them as a prerequisite for Calc III, then transferred again to a school that counted Calc III towards my degree but not I or II, compelling me to have to retake them. The good news is this is advantageous to my GPA; the bad news is... well, guess who decided to use Laplace transformations to solve petty little first-order differential equations? Me. Because BOOOOOOOORIIIIIIIING. It's an open secret in Calc II that I have paid scarcely any attention in class at all, and still set the grade scale for the rest of the class.

Here's the part where I gloat: on the last exam of the year, I got 111%. I scored higher than the grade scale permits. ("I don't always get pass tests, but when I do...")

Here's how. The grade scale in Calc II worked by first scoring all the exams by points-per-problem out of points possible. All the scores are then scaled based on the person who scored highest - if the highest score is 93%, it becomes 100% and everyone else gets (100-93)% = 7% added to their score. Then the extra credit points are factored in; every test has up to 10% worth of extra credit, so the highest theoretical score is 110%. (Did I mention this was an easy class? It was a really easy class.)

There was only one question on the last exam I missed. (NOTE: If you don't follow the rest of this paragraph until you reach the next note, that's okay.) The instructions were to integrate 1/(1+x2) from -1/2 to 1/2 (the antiderivative of 1/(1+x2) is arctanh(x) + C, AKA the inverse hyperbolic tangent, in case you haven't gotten to hyperbolic trig...). Yes, it was literally the formula exactly, so it seemed easy until I realised the calculator doesn't have an inverse hyperbolic tangent button. After mulling this over for a while, I decided I could define y = arctanh(x), solve for x, use the identity tanh(y) = (ey-e-y)/(ey+e-y), and then solve for y to learn the logarithmic equivalent of arctanh(x). (NOTE: Okay, we're back!) Good news: it worked. Bad news: I completely lost the teacher somewhere along the way, who was expecting us to solve it using Taylor series. Oops. So I lost points and got a 93%, the scale was set by someone else who got a 99%, and so after accounting for extra credit I ended up with 104%.

When I got the test back and asked about the big red question mark on that problem, teacher explained his confusion, I explained my process, teacher said OH! That makes sense, and gave me back the seven points I had lost. And, since he wasn't about to go recalculate everyone else's grades in the class, that left me with a score that was 1% beyond what was theoretically possible! :D

Okay, anecdote ends here. Those of you who ended up skipping to this point get something to read as well. Most of you are probably part of the broad portion of society who not only hate mathematics, but are openly proud of it, which is a weirdly socially-acceptable display of ignorance. Ever wonder why that is? Paul Lockhart says it's because of the way that mathematics is (erroneously) taught as a science rather than an art, and you probably do appreciate mathematics when you encounter it - you just don't recognise it when you do, because you've been told all of your life that mathematics is something totally different! You can read it here. It is a bit long, but entertaining, illuminatory, and totally worth it. Post your thoughts when you're done.

16 June 2011

...an Eternal Golden Braid

Continued from last week, when we were discussing strange loops!

Quick review: a strange loop is anything that linearly relates to itself - if you pick some one action to perform on it, you'll eventually return to exactly where you were when you began.

Now - per the last example last post, human consciousness is itself a strange loop, as shown whenever you think about thinking, or think about why you think or feel the things that you do. So it should be no surprise that philosophy, psychology, mathematics, art, and so on can all be shown to be complex, tangled hierarchies full of strange, looping structures. Here's a couple more examples, for fun:
  • Storytelling. Structures which return the characters to the place they started from at the beginning - physically, mentally, emotionally - are common throughout great literature. The Hero's Journey is a perfect example - albeit an overly specific, misleading, and rigidly stratified one. In the Iliad, the Greeks go to war; in the Odyssey, the last of them finally comes home. (In the beginning of the Odyssey, Odysseus almost gets home but fails.) There and Back Again. Don Quixote. The answers are always in the place you started, but you need the intervening book in order to recognize them.
  • Feynman diagrams. Antimatter and matter are always created together in nature, and annihilate each other when they converge again. Feynman diagrams are just how scientists keep track of their interactions. Particle creation and annihilation is happening constantly, everywhere, and usually the two particles that formed together also destroy each other shortly thereafter - the diagram looks like this (incoming energy waves convert into two particles with mass, which move apart, collide again, and re-release the energy). You'll notice, though, that the regular particle is marked with a forwards arrow, and the antiparticle is marked with a backwards arrow - because one possible interpretation of the math is that there is only one particle and it appears to sometimes be an antiparticle because at that point it has begun moving backwards in time. In other words, the particle is 'creating' and 'destroying' itself, in a single infinitesimal moment, forever.
  • Fractals. You can zoom in indefinitely and get the same image you started with, yes? Well, not for all fractals - the famous Mandelbrot Set, for instance, is unimaginably more complex than that.
  • Let's talk (briefly) about the Mandelbrot Set. Do you know the coordinate plane? With an origin and a defined axis, you can locate any point P(x, y), x units left or right and y units up or down. One of the many, many things we can do with this is to apply a formula to P to define a second point - call it P1. Let's say that if P is x units right and y units up, we'll start at the origin, and move (x squared minus y squared) units right, and (twice x times y) units up. Then, we'll move it x more units right, and y more units up, and call that P1. X and Y can be absolutely anything, and we'll always get another point... so let's say that we take what we just did to turn P into P1, and do it to P1 to get P2, then P3, P4... and so on. We'll do it forever. (Mathematicians have sneaky tricks to find out what happens if you keep doing something forever.) If we keep going forever, we'll find that either our point is now racing toward the edge of the coordinate plane, an infinite distance away - or it's still meandering about, passing by our original point P every so often. Every point that hangs about when you do this is part of the Mandelbrot Set; in this picture, these are the points in black (the non-black points are coloured based on how quickly they run for the horizon). But if the point keeps doubling back on itself when we keep doing the same thing... the Mandelbrot Set consists of all of the points that generate strange loops from a given formula. Awesome? Awesome.
Moving on.

Strange loops can be bewildering at times because our minds are adapted to a conventionally categorical, hierarchical way of thinking - the classic riddle of shallow philosophy, "the chicken or the egg", writ large. The two hands cannot actually be drawing each other, after all! That looping sort of cause and effect makes no sense to our ordinary perception.

So it's very helpful that Hofstadter addresses this. None of the loops we've discussed, he points out, are actually self-sustaining or self-perpetuating. Rather, we can restore part of our usual understanding of things by noticing that every point of the loop has both an internal cause - somewhere else in the loop - as well as an external cause, generally one single one for the loop as a whole. You stand on any step of the Penrose staircase by climbing there from a lower step - and because Penrose conceived of the staircase. Characters reach the fulfilment of their stories with the unnoticed aid of their author. A quine is produced by being run, but only because it was first run on a computer. (The bits that make up computer memory are also strange loops, tiny tiny circuit segments that continuously feed themselves their own voltage as new input.) The trigonometric functions, and fractals, and especially the Mandelbrot Set, were discovered by expanding upon math that was already known. Particle-antiparticle pairs form from errant energy waves nearby in the universe. The Liar Paradox needs someone to say it or there is no "I" to be lying and not lying.

You'll notice I skipped humans, because the human mind is more complex than any of these - it is a self-modifying strange loop. In this sentence, I am now making you aware of the fact that you are thinking about strange loops, such as this sentence and your thoughts. You've just added an additional metarecursion onto that cycle, and you can keep doing that to potentially infinite degrees, limited only by your boredom! But for the moment I'm going to tie whatever cycle you're currently on back to this sentence by noting that by thinking about that uppermost layer and the process by which your mind reflects on itself you've constructed a strange loop of strange loops. With my help, of course. This blog post provides initial external instigation. ;)

My point is, though, that your mind doesn't necessarily need the external instigation. I have no idea how many strange loops that cycle in the last paragraph, for instance, and you might have easily realised the loop of loops without my assistance. This is because your mind is a more complicated, tangled, crazy, and wonderful place than you may have ever realised before! :D

I'm not going to be able to finish this in only two posts. Next time: we have to go deeper.Link

09 June 2011

Gödel, Escher, Bach

I just realized I've never actually discussed Gödel, Escher, Bach in detail here before, and then I went and made reference to it in my last post. Shame on me. It's awesome.

Gödel, Escher, Bach is a book by Douglas Hofstadter, whose subject cannot be really concisely explained. It's about everything. At it's core, you could say it's mostly about human consciousness. And music. And math. And art. (Kind of as implied by the title. You remember Kurt Gödel, right?) And puns, and palindromes, and Möbius strips, and vinyl records that destroy the record players that play them, and the holism/reductionism dichotomy, and computer programming, and artificial intelligence, and Charles Babbage's parable of Achilles and the Tortoise, and... and, it's amazing and you should go read it. Skim the parts that are too jargony for you, if you must, but keep going through it.

Let's talk about a construct that Hofstadter introduces, called a "strange loop". It's a shorthand way to talk about things that linearly relate to themselves. It'll probably be clearer if I discuss examples instead of trying to define it:
  • You've seen M. C. Escher's lithograph of the endless stairs? (The artwork is called 'Ascending and Descending', by the way. The structure itself is called a Penrose staircase, after the fellow who actually invented it - if you can tell me who famously used the correct name, you get an imaginary cookie.) That's a strange loop. You start anywhere on the stairs, and do nothing but walk up - linearly, one direction - and you still return to the point you started from.
  • A quine is a computer program that, with no input, produces its own exact source code as output. This is a strange loop - you proceed linearly down the generations of output, and each one is still the same code with which you began.
  • The trigonometric functions are a strange loop. For the function y = sin(x), the rate at which y changes with respect to x changing is y' = cos(x). The rate at which y' changes is y'' = -sin(x). The rate at which y'' changes is y''' = -cos(x), and the rate at which y''' changes is y'''' = sin(x) = y. Applying one procedure over and over takes you back to the original result.
  • The Liar Paradox - "I am lying right now" - is a strange loop. The thought process goes, if you are lying, then that statement is a lie, so it must not be true that you are lying right now, so you must not be lying. But if you're not really lying, then when you say you are lying you must be telling the truth, in which case you really are lying. Paradox.
  • The way your mind itself works is a strange loop. This is the process called introspection - you are able to recognize that your brain is producing thoughts, and this recognition is itself one of those thoughts.
People tend to think of strange loops as tricksy, exotic, complicated, mind-blowing things, but in reality, they're everywhere. People just tend not to notice them, or to think too hard about them, when they encounter them in real life. It's both frustrating, and depressing.

I'll elaborate on the significance of this next week, but in the meantime, you can go to another fantastic strange loop by following this elegant and finely-crafted link.

07 October 2010

A Most Ingenious Solution

Answers to last week's conundrum!

Let's start where you probably started - with the simple probability that, with one opportunity to make the choice and four options to choose from, you have a 25% chance of picking any given option and therefore a 25% chance of picking the right one.

This falls apart as soon as you realize that there are two options marked 25%. Two correct options out of four gives a probability of 50%, so we'll pick that one...

...wait. There's only one answer of 50%, and one of four is 25%. Again.

This is probably where most people give up.

-----

Let's think about this a little more. Three out of our four answers are now in what amounts to a superposition of states - our argument proves that each of them is both true and false. This leaves us in a position of deciding what rule to use: either "all incompletely false statements are true" or "all incompletely true statements are false".

- If we go with the second, then we have eliminated three of the four options as acceptable answers and should default to the one remaining.
- If we go with the first, then since we have just judged that three of the four answers are true we have a 75% chance of selecting a correct answer at random.

In either case we ought to choose 75% as the correct answer!

If 75% is the correct answer, we have one chance out of four of choosing it correctly at random. Gotterdammerung!

-----

In fact, however, this is the key to solving this problem. By this chain of reasoning, we have constructed an argument that places all four answers in a superposition of states; and, depending on whether we choose "all incompletely false statements are true" or "all incompletely true statements are false", the solution is either 100% or 0%, respectively. And neither 0% or 100% is presented as a solution to choose from. This means that the best possible answer does not appear as one of our four choices, leaving us with no chance at all of choosing the correct answer! No chance at all is 0% - and so, finally, we have an answer that is permitted to be true without also being false.

Since this choice does not appear, we simply do not select any of the answers, and in so doing get the problem right.

-----

Final thought. Amidst all these convoluted circles of paradox in such a simple question, you may have forgotten that most people give up after discovering only the first circle. Having woven our way through the rest of the problem, we now learn that giving up is the best possible answer.

This means that, of all the people who ever encounter this question, most of them will answer it correctly - despite the fact that we just proved that each of them has no chance of choosing the right answer.

I'll leave you to think on that one for a while. :D

- Thursday

30 September 2010

A Most Ingenious Paradox

I received a most excellent riddle from my deranged friend Qwip yesterday, and present it to you here in modified form:

Multiple Choice: If you were to answer this question by random guessing, what is the probability that you would be correct?

(a) 25%
(b) 50%
(c) 75%
(d) 25%

Discuss!

02 September 2010

A proof about hypocrisy.

I mentioned this in my last post and present it for you here. I stole it from Raymond Smullyan, a puzzle enthusiast and recreational mathematician, and the author of What is the name of this book?, The Riddle of Scheherezade, To Mock a Mockingbird, Alice in Puzzle-Land, and The Lady or the Tiger?, among others. (His version, incidentally, is a lot shorter than mine is, because I'd like you to absorb this and not simply treat it as an interesting game. And also I run off on weird tangents sometimes. If I were to simply cut and paste it, I would already be done by now.)

Let's start with our definition of hypocrisy, so we all know exactly what we're talking about.

1: Any person who does not believe what he claims to believe is a hypocrite.

Not a lot to build on, I admit. We'll need a premise that we can agree is true, to use as a basis.

2: Everyone believes things.

I hope this is something we can agree on! Even without getting into onerous philosophical questions, I'm sure you have certain beliefs about, oh, the shape of world, or the usual color of plants in the spring, or what you had for dinner last night, or whether you will still be alive tomorrow morning. Anytime you say "I think", "I feel", or "I know", you're expressing a belief you have.

3: Any given belief is either true or false.

I'm losing a bit of accuracy here for the sake of clarity, because if I sincerely believe that colorless green ideas sleep furiously we can argue for a long time about whether that is true, false, poorly defined, or even meaningful (and, if it's not meaningful can I really believe it?) - so for the sake of getting on with it we'll assume that all beliefs can be clearly expressed in a way that is either true or false, and if there are some that can't we don't care about them in this argument anyway, so there. :P

4: Each of your beliefs is either true or false.

This follows as a syllogism from 2 and 3, so hopefully there's no disputing it.

5: You believe each of your beliefs is true.

Does this seem obvious? It's not. There's a whole convoluted discussion over Moore's Paradox, addressing sentences like "It's raining, but I don't believe that it's raining." The statement seems absurd, but there's no reason that the two halves of the statement can't both be true - maybe it's raining outside, but you're indoors and away from the windows and can't hear the drops. Moreover, both placing the situation in the past tense ("It rained, but I did not believe that it rained") and shifting the subject ("It is raining, but you don't believe it is raining") result in perfectly reasonable statements!

That conjunction must be playing tricks, yes? Actually, to somewhat oversimplify the situation, it's in the word "belief". Someone who does not believe that it's raining cannot honestly claim that it is raining, and vice versa. If our supposed speaker says that it is raining, and believes it, then they are lying when they say they believe it is not - and, as a person who lies about what they believe, is therefore a hypocrite. On the other hand, though, if they say it is raining but are truthfully claiming they don't believe it, then even if it really is raining they are still lying about the situation as they understand it! In either case, the statement indicates that the speaker is not trustworthy, either by malice or simple stupidity, regardless of whether the statement as a whole is true or false.*

Alas, I digress. The only real explanation is to understand that, when you make a statement, you are implicitly claiming to believe that statement is true. All those beliefs you've stored up in your head can be brought out whenever you like, and whenever you do, you're making a truth claim. We could create a list of all your beliefs, to make this an actuality rather than just a potentiality, but that would take far too much time.

6: At least one of your beliefs is false.

Here we come to a bit of a potential impasse, because it is not possible to prove this, universally, without actually systematically running through a list of all your beliefs and verifying them one by one. So this claim is supported merely by probability and psychology.

If you take that theoretical list of your uncountably huge number of beliefs and approach it without prior judgment - how likely is it that every single one of those beliefs is true? You can think of it as flipping a coin for each statement, if you like - tails for true and heads for false. Or, if you think your system of generating beliefs is a little hardier than that, roll a hundred-sided die for each statement, and only mark it false if you roll a 1. Even if you use a dice with twice as many sides as there are beliefs on your list because your judgment is just that sound, the odds only go down to 50-50 that there are no false statements on that list whatsoever!

If you can honestly tell me, after all that, that you are utterly confident in the truth of every single belief on that list - well, I have to honestly tell you, that's not quite delusions of godhood, but it's pretty close. Even the Pope only claims infallibility in religious matters, you know.

Oh, oh, oh, but wait! If I've just persuaded you that one of your beliefs is false, then by statement #5 you must believe that one of your beliefs is false...

7. You do not believe what you claim to believe.

That list of all your beliefs has an error on it somewhere.

Nevertheless, you continue to believe them all.

You hypocrite.

Want to join the club? There's always room for one more. :)

* There is a mathematical field called intuitionistic logic designed to resolve complications like this, by only using operations that preserve justifiability, rather than truth as in classical logic. As a result (and despite the name), it's actually stricter about what conclusions you can make than classical logic is! Read up, 'tis fascinating.

03 June 2010

Use the force

Take a slinky and stretch it out so that it's fairly taut - across a room, say. Wiggle one of the ends up and down. Notice how whenever you disturb part of it, the disturbance moves away down the length of the slinky?

This is what in physics we call a wave - a disturbance that moves. There's a whole bunch of interesting stuff that happens with it that you can play with (what happens when the disturbance reaches the end of the slinky? is a good place to start), but there's one thing in particular I want to draw your attention to.

Disturb the slinky again and pay close attention to the little loops that compose the spring. What you should notice is that while the disturbance as a whole travels away, the individual pieces of matter basically only move back and forth in line with however you disturbed it; up and down, side to side, whatever. This shouldn't be a huge surprise, right? It's not like you grabbed part of the slinky and threw it across the room or something. The particular loop of the slinky you disturbed just pulls on the neighboring loops, which pull on their neighbors, and so on; it's a function of the fact that the slinky is springy and tries to return to its original shape due to tension forces. The "wave" isn't an actual object, it's just a description of the process as a whole.

So then what happens if you set up something, like a domino chain or something, next to the far end of the slinky and then disturb the slinky sideways? You can try it if you like, though what happens is essentially what you'd expect - the wave propagates down the slinky and runs into the dominoes and knocks them over.

But we just said that the wave isn't a physical object. It's just a little disturbance in the force (heh heh); you just accomplished the same thing as if you'd thrown a baseball down to the other end of the room to knock the dominoes over, only the only thing that moved across the room is a mathematical description of the force and energy transference taking place in the slinky. No actual, physical object crossed the room at any point.

Of course, you could argue that the same thing happens if you just stretched out a chain of dominoes across the room to reach the domino on the other side, but each individual falling domino moves slightly towards the next one. There's a net motion involved in the right direction. But with the slinky, each little bit of matter only moves from side to side (if you were careful). Not only does no object cross the room, no object even moves in the right direction to cross the room.

Congratulations. You just affected an object a whole room away from you with nothing but the power of math.